Solar Energy 155 (2017) 342–353

Contents lists available at ScienceDirect

Solar Energy
journal homepage: www.elsevier.com/locate/solener

A GIS-based Fuzzy-AHP method for the evaluation of solar farms

locations: Case study in Khuzestan province, Iran
Abbas Asakereh, Mohsen Soleymani ⇑, Mohammad Javad Sheikhdavoodi
Department of Biosystems Engineering, Shahid Chamran University of Ahvaz, 61357-8315, Iran

a r t i c l e i n f o a b s t r a c t

Article history: This study was conducted to prioritize the land of Khuzestan province in Iran to install solar photovoltaic
Received 6 February 2017 farms, based on techno-economic and environmental aspects. Fuzzy logic and fuzzy membership func-
Received in revised form 13 April 2017 tions were used to create criteria layers in the environment of GIS to draw the map of the suitability
Accepted 26 May 2017
of lands. Also the Analytic Hierarchy Process (AHP) technique was used to weigh the techno-economic
and environmental criteria and to draw the final map of the suitability of lands as solar farms. Results
showed that Khuzestan province has a great potential to generate solar electricity via photovoltaic arrays.
Keywords:
Based on the results, the potential of solar electricity generation in Khuzestan through the worst scenario
AHP
Iran
is approximately 1.75 times more than the gross electricity produced in Iran in 2013. On the other hand,
Land suitability installing solar farms on the lands located at the south and the southwest of the province which are in
Photovoltaic danger of turning to desert, is a good opportunity to combat with the spread of desertification. By using
Solar power solar power facilities for multipurpose application (generating solar power and combating with deserti-
Combating desertification fication), the costs of solar power generation will be amortized and the overall cost of solar power will be
reduced. Based on the map of the suitability of the lands as solar farm, the desert and semi-desert areas of
Khuzestan province have the excellent potential of electricity generation.
Ó 2017 Elsevier Ltd. All rights reserved.

1. Introduction On the other hand, energy demand is growing rapidly by world

population and economic growth (especially in developing coun-
Energy plays a vital role in economic activities so that accessing tries), so that the existing energy resources cannot meet this
to competitive and sustainable energy resources is an essential key increasing demand. By 2050, demand for energy may be doubled
for economic growth and social development. In fact, energy con- or even tripled. So it is very essential to attain clean, renewable,
sumption per capita is one of the major index to evaluate society’s sustainable and environmentally friendly alternative energy
development (Singh, 2002). The global economy is largely resources (Mirhosseini et al., 2011).
depended on fossil energy carriers such as coal, petroleum, and Renewable energy is generally defined as the energy that is
natural gas. But the natural reservations of fossil fuels are limited obtained from resources which are naturally replenished in a short
and it is expected to be depleted within the next century, if their timescale. Nowadays and as a result of the rapid technological
consumption is continued by the current rate (Mostafaeipour development, many forms of renewable energies have become eco-
et al., 2011). Also, widespread use of fossil energy resources has nomically available. Expanding the use of renewable energy along
been seriously degrading the earth’s environment (Dickmann, with enhancing the efficiency of the energy converting devices,
2006; Shao and Chu, 2008). There are many immediate adverse may significantly lead to improve energy security, reduce climate
effects such as greenhouse gases and other polluting particles change, and earn economic benefits. Some types of renewable
imposed to the environment by the burning of fossil fuels (Li energies have unique capacity to provide cost-competitive energy
et al., 2009). Fossil fuels burning (especially incompletely) emits for remote communities (Mirhosseini et al., 2011). Also compared
a significant amount of CO2, SO2, NOx and other kinds of green- with fossil and nuclear energy, renewable energy resources have
house gases into the atmosphere (Kalogirou, 2004). the least environmental impacts. Nowadays, replacing non-
renewable energy resources with renewable ones is one of the
most important challenges and in most countries has become the
major goal in the field of energy. The growing demand of electricity
⇑ Corresponding author. and heat in the world has become the major challenge in the field
E-mail address: [email protected] (M. Soleymani). of energy and researchers are trying to deal with this challenge by

http://dx.doi.org/10.1016/j.solener.2017.05.075
0038-092X/Ó 2017 Elsevier Ltd. All rights reserved.
 A. Asakereh et al. / Solar Energy 155 (2017) 342–353 343

developing safe and sustainable energy converting systems (Fadai, or battery backup systems are very important for remote, isolated
2007). and rural areas. So solar PVs are very attractive in diverse regions
Solar energy is a type of clean and renewable energy that can be worldwide (Hernandez et al., 2014).
converted safely to the other forms of energy without emitting air Solar thermal or photovoltaic electricity without subsidies from
pollutants. In this energy conversion method, Volatile Organic the government is not still cost-competitive with fossil based elec-
Compounds (VOCs) and emissions of nitrogen and sulfur won’t tricity. The levelized cost of solar PV electricity is predicted to be
be released. Therefore it can be effectively contributed to reduce about 125 $/MWh in 2020 (Eia, 2013) and this is very comparable
the probability of occurring the acidic rain. with this cost of electricity from Conventional Coal based plants or
Although solar radiations can be utilized in a wide range of Conventional Combined Cycle Natural Gas-fired plants which will
applications, solar engineering is mainly focused on thermal pro- be about 95 and 75 $/MWh respectively. About 110 $ of the cost
cesses and photovoltaic (PV) (Asakereh et al., 2014). Solar power of electricity generating by solar PVs is capital cost (Eia, 2013).
plants convert the solar radiation into electricity, directly with But if solar power plants facilities will also be used for other pur-
PV or indirectly with concentrated solar powers (CSP). PV solar poses, these costs are amortized and reduced the overall cost of
power is one of the best options to supply the world future energy solar power. In recent years Khuzestan province has been grap-
demand sustainably (Razykov et al., 2011). The International pling with environmental problems especially desertification and
Energy Agency (IEA) has reported that solar electricity will increase micro dust crisis. Common methods of combating desertification
by 20–25% by 2050, so that it will be possible to generate are mulching, growing the deserts, increasing vegetation cover
9000 TWh electricity by PV and CSP systems. This amount of clean and so on. Photovoltaic arrays specially when used on a large scale
electricity can contribute to reduce CO2 emissions by about 6 bil- cover the land so that it doesn’t need to other measures to combat
lion tons per year (IEA, 2010). desertification.
Solar energy which is an inexhaustible and the cleanest form of Many studies have been conducted on the various fields of solar
energy with remarkable source available all over the world (Aman energy for last decade. Site selection for solar power plants and
et al., 2015; Charabi and Gastli, 2011), has been widely used by investigation the available energy in those places, was one of the
mankind for thousands of years, in different ways (light, heat and main fields of these studies (Carrion et al., 2008; Hott et al.,
power) and by applying different types of evolving and improving 2012; Li, 2013). Geographic Information Systems (GIS) has been
technologies. It is a renewable resource with a large potential to used widely in this kind of studies. Multi-Criteria Decision-
reduce dependency on limited reserves of fossil fuels and to miti- Making (MCDM) methods have been used widely as well, and
gate impacts of climate change (Shafiee and Topal, 2009). However, these methods have been integrated frequently with GIS to evalu-
related technologies are no longer cost prohibitive (Bazilian et al., ate qualitative and quantitative spatial criteria simultaneously.
2013), there are continual advances to increase the efficiency and GIS is a powerful tool which is able to handle, process and ana-
to reduce production cost. On the other hand, solar energy systems lyze a large number of spatial data and it can draw together and
can contribute to reduce greenhouse gas (GHG) emissions and air analyze data from disparate sources. It has been used frequently
pollution, to enhance energy security, to create local jobs, acceler- to select solar energy sites, at local, regional, or national scale
ate rural electrification and to improve the quality of life in remote (Arnette and Zobel, 2011). Charabi and Gastli (2011) conducted a
societies. It can be safely converted to other forms of energy with- study to specify locations suitable to stablish PV sites in Oman.
out any environmental adverse effects (Charabi and Gastli, 2011; In this study GIS was integrated with multiple criteria decision
Burney et al., 2010; Tsoutsos et al., 2005). However the construc- making methods to prioritize different regions to install PV farms
tion of solar PVs emits some GHG gasses to the environment but in Oman. The main considered criteria were availability and inten-
air pollution comes from Solar PVs during the electricity genera- sity of solar radiation, natural limitation of land, and availability of
tion is zero. Also some toxic materials are widely used in solar cells major roads and infrastructures. Maximum and minimum weight
structure. Nevertheless these environmental tolls are negligible was given to the intensity of solar radiation and availability of
compared to those of conventional energy resources (Aman et al., major roads, respectively.
2015). The analytic hierarchy process (AHP) technique along with GIS
Solar energy can be harnessed in almost all locations (sunny) was used to evaluate the environmental, locational, orographic,
and thereby it can decentralize energy supply, enhance energy and climatic factors in order to select the most suitable place to
security (IEA, 2013) and avoid from additional relevant costs install a grid-connected PV power plant in Granada, Spain
including monetary costs, transport related pollution costs, and (Carrion et al., 2008).
roading wear and tear costs. Another benefit of solar PV is the Multicriteria GIS modeling technique was used to identify suit-
reduction of water consumption which is very substantial in con- able areas for wind and solar farms in Colorado, USA. The GIS cri-
ventional power plants. Water consumption of PV in operation teria used in this study, were including distance from electricity
phase is very little and it can be preserved and be effectively uti- transmission lines, cities and roads; solar radiation potential; pop-
lized for other purposes (Aman et al., 2015). This issue is particu- ulation density and land covering condition. Each criterion had its
larly very critical for countries such as Iran that is located in the own weight to model suitable location for PV site. Mondino et al.
arid and semi-arid areas. For instance ‘‘Kazeroon Combined Cycle (2014) developed a tool in GIS environment that merges the qual-
Power Plant” in the vicinity of the ‘‘Parishan Lake” has had an itative and quantitative criteria together in order to obtain the final
important role in drying up the lake. indicator to evaluate the land suitability for large PV farm installa-
Repair and maintenance costs of solar energy facilities are very tion. Other studies have also been conducted on the field of solar
low as well. Once a solar panel is installed and worked at its max- site selection in which GIS combined with MCDM and provided a
imum efficiency there is only a little proceedings required yearly to powerful tool to merge the qualitative, quantitative and spatial cri-
ensure of remaining in working order. Also solar energy installa- teria to select suitable site (Aydin et al., 2013; Brewer et al., 2015;
tions don’t use raw materials such as oil or coal and require signif- Sánchez-Lozano et al., 2014; Sánchez-Lozano et al., 2013; Sun et al.,
icantly lower operational labor than conventional power plants 2013).
(Bhandari and Stadler, 2009). Moreover, since there are no moving Wind and solar energy are more accessible than any other
parts in solar panels structure, these parts don’t wear, live longer forms of renewable energy in the Middle East. Iran, a country
and don’t make noise at work (Timilsina et al., 2012; Ibrahim which is located in the Middle East, not only has enormous
et al., 2011). Solar PV systems can be grid-tied or off-grid. Off grid resources of fossil fuels, but also can obtain significant amount of
344 A. Asakereh et al. / Solar Energy 155 (2017) 342–353

energy from many kinds of renewable resources (Fadai, 2007). Iran zation of Iran and Renewable Energy Organization of Iran. Reliable
has the second largest energy resources in the Middle East. long-term sunshine data in Iran are rare and only available in the
Although, the latest Iranian energy balance sheets has shown a bal- regions where meteorological stations have been installed for a
ance between supply and demand of electric power, but such suc- long time. Solar insolation varies with geographical location, sea-
cess in power generation are mainly based on fossil energy sources. son, time of day, and weather conditions. However, monthly and
In order to reduce the adverse environmental effects of fossil fuels, yearly average of local insolation is relatively constant. Generally,
especially in urban areas, and also to increase energy security, gov- in large scale analysis like this study, a certain level of approxima-
ernment of Iran has planned to replace part of current using fossil tion is rather acceptable (Bergamasco and Asinari, 2011). Therefore
fuels with renewable fuels (Ghorashi and Rahimi, 2011; Ghorashi, the insolation of the regions without available data, was estimated
2007). Nevertheless, due to some economic and technical limita- by interpolation in data obtained from meteorological stations
tions, renewable energy market in Iran is still in its early stages with similar climate. The average monthly and annual amount of
and Iran energy supply is heavily depended on fossil resources. solar insolation of each station from the installation year to the
The potential of solar energy in Iran is more than the potential year of 2010, was gathered. These average values were plotted
of any other kinds of renewable energy. The Organization of on a map of studied area, based on their longitude and latitude
Renewable Energy of Iran, has done or being carried out many pro- in Universal Transverse Mercator (UTM) coordinate system, and
jects in the field of solar energy, including photovoltaic electricity then, spatial interpolation technique (Kriging interpolation) was
generation in remote areas and villages, installation of photovoltaic used for predicting insolation values for un-sampled locations,
farms, and agricultural water pumping with PV systems. Several using ArcGIS 9.3. Kriging is an advanced interpolation technique
studies also have been conducted on the field of solar energy in that estimates a value for an unmeasured location, using measured
Iran, for instance disinfection of drinking water (Mahvi, 2007), values surrounded it (Gormally et al., 2012; Ruelland et al., 2008).
using of solar energy in buildings (Alipour, 2011) and feasibility The data was provided at a resolution of 50 m in a raster format.
and potential of solar energy (Azadeh et al., 2009; Behrang et al.,
2010; Dehghan, 2011; Moghaddamnia et al., 2009; Rahimikhoob, 2.2. Criteria selection
2010; Sabziparvar, 2008). Nevertheless, Iran is in its early stages
of using solar power and only 0.0002% of generated electricity in Site selection for a PV farm is affected by some factors which
2013 was came from solar energy (Anonymous, 2013). can be classified in three main categories: Technical, Economical
Khuzestan is a province in the southwest of Iran with an area of and Environmental. These factors depend on the geographical loca-
64,019 km2 and it is located within 29°580 and 32°580 N latitude tion, the biophysical attributes and the socio-economical infras-
and 47°420 and 50°390 E longitude. The province can be basically tructure of the studied areas (Charabi and Gastli, 2011). In this
divided into two regions, mountainous in north and east, and study, the evaluation criteria were selected based on the study
plains and marshland in south and west. The climate of Khuzestan goals, spatial scale, and accessibility to the geo-referencing data-
is generally very hot and occasionally humid particularly in the base. This is why three layers were created in the environment of
south, while in winters it is mild and dry. Khuzestan population GIS, two of them (solar radiation potential and accessibility of
in 2011 was about 4.53 million people, approximately 71.0% of transport links) to consider techno-economic factors and one layer
them were living in urban areas and the rest were living in rural (human and environmental limitation) to consider environmental
areas (Statistical Centre of Iran, 2011). Khuzestan province is the limitations. Fuzzy logic and fuzzy membership functions were
largest energy genarator and the pole of crude oil extraction in used to create all layers. Each layer has a value between zero and
Iran. On the other hand, this province is the second largest electric- one. The value of zero represents the full restriction and the worst
ity consumer in the country (especially in summer days its con- suitability while the value of one shows the most suitable site to
sumption raise dramatically). Household electricity consumption install a solar farm. Finally, by using Analytic Hierarchy Process
in Khuzestan province in 2013 was 10428.4 GWh and it was the and based on the previous studies, all three layers, solar radiation
first rank among all 31 provinces in Iran (Anonymous, 2013). This potential, accessibility of transport links, and the layer of human
level of electricity consumption imposes the heavy load on the and environmental limitations were weighted and combined
power grids and power outages at the peak time of domestic elec- together to create the ultimate unique layer. Accordingly the final
tricity consumption in Khuzestan province (summer days) layer of land suitability for solar farms in Khuzestan province were
Asakereh et al., 2015. obtained.
The potential of solar energy in Khuzestan is large enough to
become the largest solar energy generator in Iran (Maleki et al., 2.2.1. Human and environmental considerations
2012). The main object of this study was to implement a special As mentioned, criteria to select suitable places for installing
decision support system (SDSS) by integrating MCDM and GIS solar farms was divided into two main categories: environmental
techniques to prioritize Khuzestan land, in order to select the most and techno-economical. Environmental criteria to install a solar
suitable site for large-scale solar farm installations in Khuzestan farm at a specific location, are the criteria which consider human
province. and environmental limitations. The places where not possible to
install solar farms, or if possible causes harmful effects on the envi-
ronment or human life, were excluded to be selected as suitable
2. Methodology locations as solar farm. These locations include cities, villages,
riparian, verge, roads, borders of protected areas, agricultural
2.1. Estimation the solar insolation lands, rivers, swamps, wetlands, pastures and so on. For more sim-
plicity and also to avoid excessive calculation, all these criteria ulti-
Solar insolation is a measure of the solar radiation received on mately were combined together and formed a raster layer to
the Earth’s surface. This value is an indicator to calculate the poten- represent the layer of human and environmental limitations (Con-
tial of solar energy generation in a given site. The greater the straint layer). These limitations will be addressed separately in the
amount of solar insolation in a given area, the greater the potential following sections. A summary of these criteria is given in Table 1.
of electricity generation (Arnette and Zobel, 2011).
The information of solar insolation of 20 stations in Khuzestan 2.2.1.1. Constraint areas. Constraint areas include water bodies,
and adjacent provinces was obtained from Meteorological Organi- ecologically sensitive areas, wildlife conservation areas, flood
 A. Asakereh et al. / Solar Energy 155 (2017) 342–353 345

Table 1
Human and environmental considerations in some studies.

Criterion Limitations Reference

Slope Slope up to 4° Sun et al. (2013)
Less than 3%: suitable between 3 and 10%: just SE to SW aspects is Bravo et al. (2007)
suitable and greater than 10%: unsuitable
Slope up to 3%, is acceptable but 1% is most economic Hang et al. (2009)
Less than 3% Carrion et al. (2008), Aydin et al. (2013)
To 3%: completely suitable 3% to 7%: Suitability is decreasing and Aydin et al. (2013)
more than 7%: unsuitable
Slope up to 10° Baban and Parry (2001), Watson and
Hudson (2015)
Slope up to 5% Gastli and Charabi (2010), Charabi and
Gastli (2011)
0–30% is the best, more than 50% is the worst Sánchez-Lozano et al. (2013)
Slope up to 15% Mondino et al. (2014)
Water body, Wetlands, River, . . . Water bodies are restricted areas Sun et al. (2013), Janke (2010), Sánchez-
Lozano et al. (2013)
Lakes, rivers, mountains and other zones considered in regional Mondino et al. (2014)
planning instruments (like zones of rivers overflow) are unsuitable
500 m buffer zone for wetlands Arnette and Zobel (2011)
400 m buffer zone for wetlands Baban and Parry (2001)
300 m buffer zone for wetlands Alabi (2010)
500 m buffer zone for rivers Arnette and Zobel (2011)
400 m buffer zone for rivers Baban and Parry (2001)
River: access zone (5 m) + patrolled area (100 m) Carrion et al. (2008)
Agriculture land Agriculture land is unsuitable Sun et al. (2013)
Non-irrigated arable land, precipitation less than or equals to Bravo et al. (2007)
400 mm/yr: Unsuitable
Cultivable land must be protected Aydin et al. (2013)
Natural reserve, protected natural areas, natural Natural reserve is unsuitable Sun et al. (2013)
reserve, all areas with environmental Wildlife designations: 1000 m buffer zone Watson and Hudson (2015), Baban and
protection Parry (2001)
All areas with environmental protection or natural reserve are Bravo et al. (2007), Sun et al. (2013),
restricted areas Mondino et al. (2014), Carrion et al.
(2008)
1000 m buffer around conservation areas Arnette and Zobel (2011)
Residential areas Urban: maximize (1–5 km) Aydin et al. (2013)
Residential areas: 500 m buffer zone Watson and Hudson (2015), Baban and
Parry (2001)
Roads and railroad network Road: 500 m buffer zone Al-Yahyai et al. (2012)
Road: 150 m buffer zone Dunsford et al. (2003)
Roads and railroad network areas: restriction Sánchez-Lozano et al. (2013)

1
areas, village and city boundary, roads, railroad and areas with
steep slope. One of the main criteria applied to determine the 0.8
potential of a location to install a solar farm, is the current land
use. The allocated land for solar farms, cannot be used for other 0.6
purposes. As a result, only barren land or land with poor vegetation µ
0.4
cover which is not suitable for other useful purposes should be
considered as potential land for solar energy farms. In study in 0.2
the greater southern Appalachian Mountains (USA), land with veg-
etation cover less than 15% was considered as appropriate site for 0
0 0.5 1 1.5 2
solar farm installation (Arnette and Zobel, 2011), but in Colorado,
Distance from agricultural areas (km)
short vegetation, logged areas, or barren land, that cannot reduce
solar insolation, were considered as ideal locations (Janke, 2010). Fig. 1. Limitation in term of agricultural areas.
Large scale installations of PV systems may damage the arable
land (Tsoutsos et al., 2005). On the other hand, protection of arable
land is one of the main environmental objective. Thus fertile arable (l = 0), land inside the buffer zone is partially suitable (0 < l < 1)
land mustn’t be used for solar farms. Based on Aydin et al. (2013), and land outside the buffer zone (l = 1), in the absence of other
agricultural areas must be excluded from being selected as solar barriers and restrictions, is completely suitable for this purpose.
farms. For greater certainty a buffer zone with radius of one km Rivers, wetlands, lakes and wildlife conservation areas are the
should be considered around these areas. That is, the acceptability sensitive areas that must be kept away from toxic materials. On
of land for solar power plant within the buffer zone increases from the other hand, PV modules consist of some toxic and hazardous
zero to one, respectively for the border of a farmland and a distance materials that must be avoided to be released into the environ-
of one km from that border (Aydin et al., 2013). Fig. 1 shows the ment. Therefore solar farms should be installed at a safe distance
fuzzy membership value of land (l) for solar farm installation, from the scope of these sensitive areas. This distance from rivers,
based on the distance from agricultural areas. This figure shows wetlands, lakes and wildlife conservation areas was different in
that the agricultural land is excluded to be selected as a solar farm previous studies (Table 1). So based on previous studies (Table 1),
346 A. Asakereh et al. / Solar Energy 155 (2017) 342–353

a fuzzy membership function (Eq. (1)) was designed to consider with short vegetation such as shrubs, prairies, grasslands, scrub
this restriction: and steppe, were considered as ideal locations for solar farm instal-
8 lation by Janke (Janke, 2010). The fuzzy membership value for low-
>
< 0 x 6 100
dense, semi-dense and dense rangelands, was considered to be 1,
lRWC ¼ x100
100 6 x 6 400 ð1Þ 0.5 and 0, respectively.
> 300
:
1 x P 400 Iran is a country with poor vegetation cover and protection
from this vegetation cover especially in Khuzestan province where
where: x, is the distance from rivers, wetlands and conservation
desertification and micro-dust phenomena has been caused many
areas in meter; and lRWC ; is the fuzzy membership value of land
environmental problems in the recent decade, is very essential.
for solar farm installation (coefficient of rivers, wetlands and con- One of the important type of land cover in Khuzestan province is
servation areas). shrubbery and reed-bed of water or swamp margins. These areas
Acceptability of land around the lakes was determined by Eq. not only are the habitat of various birds and animals, but also play
(2) as follows: an important role in combating desertification. Eq. (6) shows the
8
> x 6 300 fuzzy membership functions of acceptability of land as solar farm,
< 0
considering shrubberies and reed-bed. These function was created
lL ¼ x300
300 6 x 6 500 ð2Þ
> 200
: based on the data of Table 1.
1 x P 500 8
>
< 0 x 6 100
where: lL , is the fuzzy membership value of land around the lakes
lY ¼ x100
100 6 x 6 400 ð6Þ
for solar farm installation (coefficient of lakes). > 300
:
Clearly, it is not possible to install a large PV farm among the 1 x P 400
buildings. Also to install a solar power plant, economic issues
Roads, railroads, sand dune, and flood zones are the areas where
related to the other profitable uses of land should be considered.
not possible or not economic to install a large solar farm. There are
For instance, land near urban areas may be more profitable for
different types of roads and railroads in Khuzestan province. In Iran
other commercial applications than to allocate to a solar farm.
like many countries, depending on the type of the roads, any con-
Eqs. (3) and (4) were designed based on Table 1, to exclude land
struction from the centerline of roads and railroads to a certain dis-
near urban and rural areas from selecting as the potential land
tance, is prohibited. This distance, besides 100 m (as buffer zone)
for solar farms:
was considered as non-exploitable area and was excluded from
8
>
< 0 x 6 1000 being selected as potential area to install a solar farm. Also, the fol-
lU ¼ x1000
1000 6 x 6 5000 ð3Þ lowing fuzzy set (Eq. (7)) was applied to consider the limitation of
> 4000
: flood zones:
1 x P 5000
8
8 >
< 0 x 6 100
>
< 0 x 6 300 lF ¼ x100
100 6 x 6 400 ð7Þ
> 300
:
lW ¼ x300
300 6 x 6 700 ð4Þ 1 x P 400
> 400
:
1 x P 700
Slope is a topographic feature which can strongly affect the pro-
where: lU and lW are the fuzzy membership values of urban and ject costs. So to select a proper location for a solar power plant, the
rural areas, respectively. slope of that location can play an important role. In a steep land,
As mentioned before, it is better to install solar farms on land panels can create the shadow over the neighboring panels and
with poor vegetation cover. Forest areas are not applicable to hereby the efficiency of energy conversion decreases. Also setting
install solar farms, because of shading of trees on collectors, photo- up the infrastructure in a flat land is a lot easier and as a result,
voltaic modules and other devices. Forest areas also must be pro- the overall construction costs decrease. In a flat terrain, on the
tected from possible harmful effects of solar farms. The forests of other hand, changes such as excavation or embankment and conse-
Khuzestan province can be divided into three categories: dense quently manipulation of the natural landscape can be minimized.
(forest with more than 50% canopy cover), semi-dense (forest with As shown in Table 1, different maximum slope are recommended
25–50% canopy cover) and low-dense (forest with 5–25% canopy for a PV farm installation.
cover). Different minimum safe distances from forests have been The Digital elevation model (DEM) map of Khuzestan Province
considered in different studies (Baban and Parry, 2001; Yue and with a pixel of 10 m was obtained from Iran National Cartographic
Wang, 2007). Based on similar studies and the condition of Khuzes- Center. The value of each pixel represents the elevation which can
tan forests, the possibility of land next to the forests to select as be used to calculate slope with ArcMap.
solar farm is represented by a fuzzy set, as Eq. (5): So fuzzy membership functions were determined (Eq. (8))
8 based on the maximum gradients expressed in previous studies.
>
< 0 x 6 100 The value of one was given to the cells (in raster format) with a
lF ¼ x100
100 6 x 6 500 ð5Þ slope of 0–3%, and the fuzzy value of the cells with a slope between
> 400
:
1 x P 500 3 and 10% were reduced from one to zero respectively. Ultimately
the area with slop more than 10% was considered as non-
This function indicates that dense forest areas and 100 m buffer exploitable areas.
zone around them is considered as non-exploitable area, and land 8
acceptability for PV farms increases from zero to one respectively >
< 1 x63
from 100 m to 500 m next to the dense forest area. Also, semi lS ¼ x3
3 6 x 6 10 ð8Þ
dense forest areas were considered as non-exploitable areas but > 7
:
0 x P 10
the buffer zone was not considered for them. Ultimately the fuzzy
membership value for low-dense forest areas, was considered to be Finally each mentioned criterion was modeled as a layer (sub-
0.5. layers) and was drawn as a map (in raster format) in GIS environ-
Rangelands of Khuzestan province, as forests, can be divided ment, with 50-m spatial resolution. Ultimately, all layers were
into three categories: dense, semi-dense and low-dense. Areas multiplied together and formed the final restriction layer. Loca-
 A. Asakereh et al. / Solar Energy 155 (2017) 342–353 347

tions with value of zero in each limitation layer also would be zero in the range of [0, 6] kWh/m2/day change in the interval of [0, 1].
in the final restriction layer. For example, in a location with the Solar intensity of 4 kWh/m2/day was valued by 0.5 (moderate).
high risk of flood where it is completely unsuitable for a solar farm Proximity to the transport links (roads and railroads) is another
(fuzzy membership value is zero), the final restriction layer is com- essential factor that determines the level of difficulty to carry PV
pletely unsuitable and fuzzy membership value will be zero too, cells into the site. It is easier to establish PV utilities at locations
even if fuzzy membership value of all other layers be one. Loca- close to the roads or railroads. Proximity to the roads or railroads
tions with value of one in all sublayers, also have the value of avoids additional cost of infrastructure construction and conse-
one in the final restriction layer. Since the other locations in sub- quently reduces probable damages to the environment (Brewer
layers have the fuzzy membership values between zero and one, et al., 2015). The maximum acceptable distance from the roads
so the fuzzy membership value of the final restriction layer in these and railroads depends on the circumstances of each region. So dif-
locations will be lower than that of the sublayers. Locations in the ferent distances are recommended in different studies (for exam-
final restriction layer which have the fuzzy membership value of ple: 10 km Baban and Parry, 2001, 40 km (Asakereh et al., 2014;
one are the most suitable locations without any limitation to Dahle et al., 2008) or 80 km (Assessing the Potential for
install a solar farm. Renewable Energy on Public Lands, 2003). Some studies haven’t
considered the maximum suitable distance from the roads or rail-
roads. These studies valued the locations suitability based on the
2.2.2. Techno-economic criteria distance of that location from the roads, so that the worst value
The amount of solar radiation received on the earth’s surface is assigned to the farthest locations and the best value is assigned
determines the amount of solar energy can be converted to the to the nearest locations from the roads or railroads (Brewer et al.,
electricity. The condition of the solar radiation in a location is a 2015; Charabi and Gastli, 2011; Janke, 2010; Sánchez-Lozano
key indicator to qualify that location to install a solar power plant. et al., 2014; Watson and Hudson, 2015).
If all criteria are in good condition except the solar radiation, that All these recommendations were considered to define a fuzzy
location will not be qualified as a good location to install a solar membership function to determine Khuzestan land proximity to
power plant. US National Renewable energy Laboratory (NREL) the transport links, as Fig. 3.
classified solar radiation qualities into four categories including:
moderate (less than 4 kWh/m2/day), good (4–5 kWh/m2/day), very
good (5–6 kWh/m2/day) and excellent (>6 kWh/m2/day) 2.3. AHP
(Phuangpornpitak and Tia, 2011). Based on Aydin et al. (2013),
generation of solar energy economically, requires solar radiation The aim of a MCDM method is to investigate a number of alter-
at least 4.5 kWh/m2/day. This amount is slightly different in vari- natives in the light of multiple criteria and conflicting objectives
ous studies and regions (for example: 4.19 kWh/m2/day in the (Sánchez-Lozano et al., 2013). AHP is one of the well-known MCDM
south of USA (Arnette and Zobel, 2011) and 5 kWh/m2/day in the method invented by Saaty in 1970s as a decision making tool to
south-east of Spain (Sánchez-Lozano et al., 2013). The minimum resolve unstructured problems (Saaty, 1980). It is the powerful
annual solar radiation in Oman was calculated to be 1522 kWh/ and useful MCDM approach tool for dealing with complex decision
m2/year (4.17 kWh/m2/day) and it was found too low to be suitable problems. AHP can handle the decision making problems which
for a solar PV development (Charabi and Gastli, 2011). The mini- require a high degree of flexibility and reliability (Carrion et al.,
mum and maximum solar radiation in England was calculated to 2008). AHP is based on the pairwise comparisons. In this method,
be 536 (1.47 kWh/m2/day) and 1076 (2.95 kWh/m2/day) kWh/ a decision-maker forms a hierarchical decision tree and determines
m2/year, respectively (Watson and Hudson, 2015). Brewer et al. its indices and options. It provides a comprehensive and rational
(2015) conducted a study on the site suitability for large-scale solar framework for structuring a decision problem, for representing
power installation in the southwest of the United States and found and quantifying its elements, for relating those elements to overall
that the minimum and maximum common radiation was 3 and goals and for evaluating alternative solutions.
8 kWh/m2/day respectively. AHP is a technique that widely used in studies which is apply-
According to the classification of National Renewable energy ing GIS-MCDM methods in the field of sustainable energy (Pohekar
Laboratory and based on similar studies, a fuzzy membership and Ramachandran, 2004). In this study, it is found that the com-
equation was created by using of Gaussian function, to determine bination of GIS and AHP will be useful to select the suitable loca-
the solar radiation potential of Khuzestan province (Fig. 2). In this tion for a PV solar farm. The basic theory of AHP may be
equation, locations with 6 kWh/m2/day and more was valued by 1 summarized as follows (Saaty, 2008):
(Excellent for solar farm installation, provided that other suitable
conditions) and the values of the locations with the solar intensity
1
0.9
1
0.8
0.7
0.8
0.6
0.5
µ

0.6
0.4
µ

0.4 0.3
0.2
0.2
0.1
0
0
0 1 2 3 4 5 6 7 0 5 10 15 20 25 30 35 40 45 50
Solar energy potential (kWh/m2/day) Distance from transport links (km)
Fig. 2. Fuzzy set for solar energy potential. Fig. 3. Fuzzy set for transport links adjacency.
348 A. Asakereh et al. / Solar Energy 155 (2017) 342–353

1. Modeling the problem as a hierarchy which is containing the efficiency (which include efficiency resulting in dust and dirt on the
decision goal, the alternatives to reach this goal, and the num- PV array (94%), DC electrical panel (98%), inverter efficiency (90%)
ber of criteria to evaluate these alternatives. and reflection of the sunbeams of the array (97%)). The value of
2. Defining priorities among the elements of the hierarchy by ginst was estimated 80%. PV module efficiency is expressed at the
making a series of judgments based on pairwise comparisons Standard Test Conditions (STC) which means at 1000 W/m2 of solar
of the elements. The values used in the pair-wise comparison, irradiance, an air mass of 1.5 spectrum and with a module temper-
as shown in Table 2, are in the interval of 1–9. ature of 25 °C. The efficiency of PV modules has been improving
3. Synthesize these judgments to yield a set of overall priorities for under the continuous development of PV module technology. Dif-
the hierarchy. ferent module efficiencies have been applied in similar studies
4. Check the consistency of the judgments. (Charabi and Gastli, 2011; Bergamasco and Asinari, 2011;
5. Come to a final decision based on the results of this process. Gormally et al., 2012; Wakeyama and Ehara, 2010). Therefore three
scenarios were considered based on three efficiency values of 10, 15
2.4. Solar energy conversion and 20% (Table 3).
According to Bergamasco and Asinari (2011), efficiency reducing
The annual potential of solar electricity generation, at a selected over the lifetime of the modules is negligible and can be ignored.
site can be calculated based on the average annual solar radiation Also the loss of efficiency resulted from the temperature variations
intensity, site area and the efficiency of solar energy conversion is about 10% (gT = 0.9). Module installation angle is the angle of the
system. Eq. (1) was used to calculate the potential of solar electric- panel toward the south (in the northern hemisphere). To consider
ity generation (Charabi and Gastli, 2011).
GP ¼ SR  CA  AF  gTot
9
where: GP is the annual potential of solar electricity generation Max
8
(kWh/year), SR is the annual average solar radiation intensity
(kWh/m2/day), CA is the total area of the selected site (m2), AF is 7 Min
the area factor and g is the efficiency of solar energy conversion 6
kWh/m2/day

system. AF shows what fraction of the site area can be covered by

5
solar panels. The value of AF was selected based on the maximum
land can be covered by PV panels with minimum shading effect. 4
The value of AF in a similar study at the greater southern Appala- 3
chian mountains (in USA) Arnette and Zobel, 2011 and Oman
2
(Charabi and Gastli, 2011) was considered to be 75% and 70%
respectively. 1
One of the most important technical issues in solar electricity 0
generation is the efficiency of PV system that converts the solar Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
energy into the electrical energy. The total system efficiency is
Fig. 4. Maximum and minimum solar insolation for each month of the year.
influenced by the module efficiency, losses of dust and dirt on
the PV array, DC electrical panel, temperature variations, module
installation angle, inverter efficiency and the reflection of the sun-
beams from the array. Considering all above factors, the total sys-
tem efficiency (gTot ) is calculated by the following equation
(Bergamasco and Asinari, 2011):
gTot ¼ gMod  gT  gAZ  gInst
where: gMod , is the module efficiency; gT , temperature variations;
gAZ , module installation angle efficiency and gInst is the installation

Table 2
The values that used are in the pair-wise comparison.

Verbal judgements of preferences between alternative Numerical rating

Extremely preferred 9
Very strongly preferred 7
Strongly preferred 5
Moderately preferred 3
Equally preferred 1
Intermediate values 2, 4, 6, 8

Table 3
Total system efficiency for the modules with three efficiencies.

H A B C
gMod 0.10 0.15 0.20
gTH 0.90 0.90 0.90
gAZ 0.90 0.90 0.90
gInst 0.80 0.80 0.80
gTot 0.0648 0.0972 0.129
Fig. 5. Fuzzy value of solar energy potential.
A. Asakereh et al. / Solar Energy 155 (2017) 342–353 349

Fig. 6. Constraint fuzzy layers.

350 A. Asakereh et al. / Solar Energy 155 (2017) 342–353

reducing the efficiency of photovoltaic system due to the incorrect the plains of the province cause the greatest constraints to install
angle of the installation, gAZ was introduced. So this coefficient was solar power plants. About 24.46% of all land in Khuzestan province
assumed 0.9 (gAZ =90) based on Bergamasco and Asinari (2011). have a slope more than 10% which is unsuitable to install a solar
Total system efficiency of the modules with three efficiencies are farm and its fuzzy membership value was considered to be zero.
shown in Table 3, where gTot is the total efficiency of the system. Land with a slope of 3% or less is completely suitable for solar farm
installation and its fuzzy membership value was considered to be
one. This type of land covers 61.65 of the area of the province. Only
3. Results and discussions
13.89% of the province land has a slope between 3 and 10%.
About 21% of the total area of the province is fertile agricultural
3.1. Solar energy potential
land. So it is unsuitable for a solar farm and the fuzzy membership
value of this type of land was considered to be zero. In this point of
The results of kriging interpolation of annual solar insolation,
view, 63% of the total land of Khuzestan with the fuzzy member-
with an approximate resolution of 50 m, were plotted in Arc GIS
ship value of one is completely suitable for solar farm installation
9.3. The amount of solar insolation ranges from 5.09 to
and the others are relatively suitable with the fuzzy membership
5.49 kWh/m2/day annually, which following to the classification
value between zero and one. Urban and rural infrastructures stand
of NREL, belongs to the class of very good for solar electricity
in the fourth place of restrictions and 9.42% of province area is non
generation. Therefore, apart from other constraints such as
exploitable (fuzzy value equal to zero) because of this type of con-
techno-economic or environmental limitations, all areas of
straint. Also the condition of other restrictions such as rivers, wet-
Khuzestan province can be potentially used as much suitable solar
lands, lakes, conservation and protected areas, rangelands, shrub
farm site (Phuangpornpitak and Tia, 2011) (Fig. 4). Fig. 4 shows the
lands, reed bed, roads and railroads, sand dune, and flood zones
monthly solar insolation in Khuzestan province. The highest
are shown in Fig. 6.
amount of solar insolation which is greater than 7 kWh/m2/day
Totally, 71.47% of the areas of Khuzestan province with the
occurs during the summer months. Generally, the average solar
fuzzy value of zero is non exploitable and in return 8.32% is com-
insolation in the province increases from January to June and then
pletely exploitable with the fuzzy value of one. The rest has fuzzy
reduces to Dec (Fig. 4). Maximum insolation intensity which is
value between zero and one. The most exploitable areas are located
approximately 7.73 kWh/m2/day occurs in Jun. On the other hand
in plain and flat land, generally in the southern part of the pro-
the maximum electricity consumption in Khuzestan province,
vince. Installation of solar power plants in these areas, by covering
occurs during the hot summer simultaneously with this maximum
earth surface can prevent the creation and expansion of deserts.
insolation intensity (April to September). Average solar insolation
Total annual solar insolation of locations with the fuzzy value of
in these months is at least 6 kWh/m2/day. Therefore it is the best
one and between zero and one is about 10,105 and 24,658 TWh,
fortune to alleviate the load on the power grid and to reduce the risk
respectively. Assuming the efficiency of 6.48% for energy convert-
of blackouts, by generating solar electricity as a good alternative to
ing system and regardless of techno-economic aspects, the poten-
the current fossil based electricity. The intensity of solar insolation
tial of electricity generation in areas with the fuzzy value of one
in the southern areas of Khuzestan province in seven months of the
(without any constraint), will be 458.4 TWh/year that is about 15
year (Mar to Sep) is higher than 5 kWh/m2/day which is very good
times more than the electricity consumption of the province or
to generate electricity based on the NREL classification. Minimum
approximately 1.75 times more than the gross electricity produced
solar insolation is about 2.64–3.10 kWh/m2/day and occurs in
December.
A map based on the fuzzy membership values of solar energy
potential was plotted on the map of Khuzestan province in raster
format with a 50 m resolution (Fig. 5). The fuzzy membership val-
ues of solar energy potential, range from 0.866 to 0.956, and are
increasing from the north to the south. The fuzzy membership val-
ues of 42.30% of land of Khuzestan province is greater than 0.9.
Therefore it is obvious that all areas of Khuzestan province have
the great theoretical potential for solar energy generation but
due to various constraints and low efficiency of solar energy con-
version systems, only a small portion of this potential can be
harnessed.
In recent years Khuzestan province has been grappling with
environmental problems especially desertification and micro dust
crisis. Although the majority of micro-dusts comes from the neigh-
boring country, Iraq and is originated from dried marshes of rivers
Tigris and Euphrates (Hamidi et al., 2013), but there are several
internal active hotspots on the South and West of the province.
By installing solar farms in these areas where there is a great
potential of solar energy (Fig. 5), it is not only possible to generate
a great deal of renewable and sustainable energy but also to com-
bat with desertification crisis in Khuzestan province by covering
land with PV arrays.

3.2. Human and environmental constraints

Fig. 6 shows the fuzzy layers of all constraints. As this figure

shows, many restrictions are overlapping. Steep terrains and
forested areas in the mountainous areas and agricultural land in Fig. 7. Access to transport links in the province.
 A. Asakereh et al. / Solar Energy 155 (2017) 342–353 351

in Iran in 2013. Total annual gross electricity production in Iran eliminated at first. The layers, then were combined based on their
was 262433.2 GWh in 2013 (Anonymous, 2013). estimated weights in Table 4. Ultimately, as shown in Fig. 8, the
final layer which is determining the land suitability for PV farms
3.3. Availability of transport links installation was created in raster format based on AHP methods.
Changing the color on the map reflect the variation of fuzzy values
The fuzzy layer of the access to the transport links in Khuzestan which vary from zero to 0.976 and represent non-exploitable loca-
province is shown in Fig. 7. Almost all areas of the province have tions and the best locations to install solar farms respectively.
good access to the roads or railroads so that about 84% of the land Five suitability levels were defined based on the fuzzy member-
of the province has a fuzzy value more than 0.9 and only 0.57 of the ship values of locations, namely unsuitable, poor, moderate, good
province was non-exploitable (fuzzy value was zero). and excellent with the fuzzy membership value of [0, 0.6], (0.6,
Another important techno-economic indicator is availability of 0.7], (0.7, 0.8], (0.8, 0.9] and (0.9, 0.976], respectively. Fig. 8(b)
electricity transmission lines. But unfortunately the map of these shows the condition of this classification on the map of Khuzestan
lines was not publicly available. So it was no possibility to review province. It should be emphasized again that the high potential of
this factor. solar energy in desert areas or areas with high risk of desertifica-
tion (South and West regions of the province) provides a good
3.4. Determining suitable land opportunity to combat with desertification by installing solar
farms on these areas and covering land surface with photovoltaic
Finally the three main criteria (solar energy potential, environ- arrays.
mental constraints and availability of transport links) were com- Fig. 8 shows that only a small portion of the province area
pared pair-wise and the weight of each criterion was estimated (8.87%) has an excellent potential to install solar farms. However
according to the guidelines of AHP method (Table 4). The impor- these areas are quite sufficient to generate large amounts of solar
tance and weight of each criterion was selected based on the pro- energy. Areas with good and moderate suitability level are
fessional experts (the results of similar studies). Therefore the included 4.98 and 9.87% of total areas, respectively. The potential
greatest weight was given to the solar energy potential as the most of solar electricity generation, using PV systems with different effi-
important criterion, while the lowest weight was given to the ciencies of modules is presented in Table 5. The potential of elec-
availability of transport links as the least important criterion. tricity generation from the areas with excellent suitability level,
Table 4 shows the result of estimated weights for each one of three assuming 6.48% for PV system efficiency, will be 480 TWh/year.
main criteria. This amount is 1.83 times more than the total annual gross elec-
For determining suitable sites, non-exploitable areas which tricity produced in Iran and also 7.46 times more than household
have fuzzy value of zero in sublayers (solar energy potential, envi- electricity consumption in Iran in 2013. Also the cumulative poten-
ronmental constraints and availability of transport links) were tial of electricity generation from good and excellent suitability

Table 4
Pair-wise comparison matrix of objectives and estimated weights.

Objective Solar energy potential Environmental constraints Transport links adjacency Weight
Solar energy potential 1 1.95 3 0.539
Environmental constraints 0.513 1 1.8 0.291
Availability of transport links 0.333 0.556 1 0.170

Fig. 8. Land suitability for solar farm installation.

352 A. Asakereh et al. / Solar Energy 155 (2017) 342–353

Table 5
Solar electricity potential by using PV system with different module efficiencies in different classes.

Class name Unsuitable Poor Moderate Good Excellent

Area of province (Percent) 71.47 4.90 9.87 4.98 8.78
Electricity Generation Potential (TWh/year) Aa 3904 258 530 271 480
Bb 5856 388 795 406 721
Cc 7772 515 1055 539 956
a
The efficiency of the system of solar energy conversion, 6.48% (the efficiency of panel, 10%).
b
The efficiency of the system of solar energy conversion, 9.72% (the efficiency of panel, 15%).
c
The efficiency of the system of solar energy conversion, 12.9% (the efficiency of panel, 20%).

levels, in the same conditions, will be 2.86 times more than the Baban, S.M.J., Parry, T., 2001. Developing and applying a GIS-assisted approach to
locating wind farms in the UK. Renew. Energy 24, 59–71.
total annual gross electricity produced in Iran in 2013. Therefore
Bazilian, M., Onyeji, I., Liebreich, M., MacGill, I., Chase, J., Shah, J., Gielen, D., Arent,
it is clear that there is a huge potential for electricity generation D., Landfear, D., Zhengrong, S., 2013. Re-considering the economics of
using solar farms in Khuzestan province. photovoltaic power. Renew. Energy 53, 329–338.
Behrang, M.A., Assareh, E., Ghanbarzadeh, A., Noghrehabadi, A.R., 2010. The
potential of different artificial neural network (ANN) techniques in daily
global solar radiation modeling based on meteorological data. Sol. Energy 84,
4. Conclusion
1468–1480.
Bergamasco, L., Asinari, P., 2011. Scalable methodology for the photovoltaic solar
In this study, the suitability of the lands of Khuzestan province energy potential assessment based on available roof surface area: application to
Piedmont Region, Italy. Sol. Energy 85, 1041–1055.
in Iran has studied regarding to the techno-economic and environ-
Bhandari, R., Stadler, I., 2009. Grid parity analysis of solar photovoltaic systems in
mental aspects. Results showed that Khuzestan province has a Germany using experience curves. Sol. Energy 83 (9), 1634–1644.
great potential to generate solar electricity via photovoltaic arrays. Bravo, J.D., Casals, X.G., Pascua, I.P., 2007. GIS approach to the definition of capacity
Based on the results, the potential of electricity generation of the and generation ceilings of renewable energy technologies. Energy Policy 35,
4879–4892.
province land in the worst case scenario is more than the gross Brewer, J., Ames, D.P., Solan, D., Lee, R., Carlisle, J., 2015. Using GIS analytics and
electricity produced in Iran. On the other hand, installing solar social preference data to evaluate utility-scale solar power site suitability.
farms on the lands located at the south and the southwest of the Renew. Energy 81, 825–836.
Burney, J., Woltering, L., Burke, M., Naylor, R., Asternak, D., 2010. Solar-powered drip
province which are in danger of turning to desert, is a good oppor- irrigation enhances food security in the Sudano-Sahel. Proc. Natl. Acad. Sci. U. S.
tunity to combat with the spread of deserts. Based on the map of A. 107 (5), 1848–1853.
the suitability of the lands as solar farm, these areas have the Carrion, J.A., Estrella, A.E., Dols, F.A., Toro, M.Z., Rodriguez, M., Ridao, A.R., 2008.
Environmental decision-support systems for evaluating the carrying capacity of
excellent potential of electricity generation. land areas: optimal site selection for grid-connected photovoltaic power plants.
Renew. Sustain. Energy Rev. 12, 2358–2380.
Charabi, Y., Gastli, A., 2011. PV site suitability analysis using GIS-based spatial fuzzy
Acknowledgements multi-criteria evaluation. Renew. Energy 36, 2554–2561.
Dahle, D., Elliott, D., Heimiller, D., Mehos, M., Ro-bichaud, R., Schwartz, M., Stafford,
This research was supported by Shahid Chamran University of B., Walker, A., 2008. Assessing the Potential for Renewable Energy Development
on DOE Legacy Management Lands. National Renewable Energy Laboratory.
Ahvaz. The authors also would like to thank Iran Meteorological Dehghan, A.A., 2011. Status and potentials of renewable energies in Yazd province-
Organization, Renewable Energies Organization of Iran (SUNA), Iran. Renew. Sustain. Energy Rev. 15, 1491–1496.
Jihad-e Agriculture Organization of Khuzestan province and Iran Dickmann, D.I., 2006. Silviculture and biology of short rotation woody crops in
temperate regions then and now. Biomass Bioenergy 30, 696–705.
National Cartographic Center for providing the data for this Dunsford, H., Macfarlane, R., Turner, K., 2003. The Development of a Regional
research study. Geographical Information System for the North East Renewable Energy
Strategy. Centre for Environmental and Spatial Analysis. University of
Northumbria.
References Eia, U., 2013. Levelized Cost of New Generation Resources in the Annual Energy
Outlook 2011. US Energy Information Administration, Washington DC.
Alabi, O.O., 2010. An investigation on using GIS to prospect for renewable energy in Fadai, D., 2007. Utilization of renewable energy sources for power generation in
Nigeria (Phd Thesis). University of Missouri-Kansas City. Iran. Renew. Sustain. Energy Rev. 11, 173–181.
Alipour, V., 2011. A Test Reference Year for Ahvaz, Iran (M.Sc Thesis). University of Gastli, A., Charabi, Y., 2010. Solar electricity prospects in Oman using GIS-based
Manchester, UK. solar radiation maps. Renew. Sustain. Energy Rev. 14, 790–797.
Al-Yahyai, S., Charabi, Y., Gastli, A., Al-Badi, A., 2012. Wind farm land suitability Ghorashi, A.H., 2007. Prospects of nuclear power plants for sustainable energy
indexing using multi-criteria analysis. Renew. Energy 44, 80–87. development in Islamic Republic of Iran. Energy Policy 35, 1643–1647.
Aman, M.M., Bakar, A.H.A., Solangi, K.H., Hossain, M.S., Badarudin, A., Jasmon, G.B., Ghorashi, A.H., Rahimi, A., 2011. Renewable and non-renewable energy status in
Mokhlis, H., Kazi, S.N., 2015. A review of Safety, Health and Environmental Iran: Art of know-how and technology-gaps. Renew. Sustain. Energy Rev. 15,
(SHE) issues of solar energy system. Renew. Sustain. Energy Rev. 41, 1190– 729–736.
1204. Gormally, A.M., Whyatt, J.D., Timmis, R.J., Pooley, C.G., 2012. A regional scale
Anonymous, 2013. Energy balance sheet, Iran Ministry of Energy Deputy of assessment of local renewable energy resources in Cumbria, UK. Energy Policy
Electricity and Energy Affairs. 50, 283–293.
Arnette, A.N., Zobel, C.W., 2011. Spatial analysis of renewable energy potential in Hamidi, M., Kavianpour, M.R., Shao, Y., 2013. Synoptic analysis of dust storms in the
the greater southern Appalachian mountains. Renew. Energy 36, 2785–2798. Middle East. APJAS 49 (3), 279–286.
Asakereh, A., Omid, M., Alimardani, R., Sarmadian, F., 2014. Developing a GIS-based Hang, Q., Jun, Z., Xiao, Y., Junkui, C., 2009. Prospect of concentrating solar power in
Fuzzy AHP model for selecting solar energy sites in Shodirwan Region in Iran. China – the sustainable future. Renew. Sustain. Energy Rev. 12, 2505–2514.
IJAST 68, 37–48. Hernandez, R.R., Allen, E.B., Easter, S.B., Murphy-Mariscal, M.L., Maestre, F.T.,
Asakereh, A., Omid, M., Alimardani, R., Sarmadian, F., 2015. Investigating potential Tavassoli, M., Barrows, C.W., Belnap, J., Ochoa-Hueso, R., Ravi, S., Allen, M.F.,
of wind energy in Mahshahr, Iran. WE 39 (4), 369–384. 2014. Environmental impacts of utility-scale solar energy. Renew. Sustain.
Assessing the Potential for Renewable Energy on Public Lands, 2003. U.S. Energy Rev. 29, 766–779.
Department of the Interior Bureau of Land Management and U.S. Department Hott, R., Santini, R., Brownson, J., 2012. GIS-base spatial analysis for large-scal solar
of Energy Efficiency and Renewable Energy. http://www.osti.gov/bridge power and transmission line issues: case studt of Wyoming. In: U.S. Proceedings
(accessed 15.03.02). of the 41st American Solar Energy Society Meeting, May 13–17; 2012.
Aydin, N.Y., Kentel, E., Duzgun, H.S., 2013. GIS-based site selection methodology for Ibrahim, A., Othman, M.Y., Ruslan, M.H., Mat, Sohif, Sopian, K., 2011. Recent
hybrid renewable energy systems: a case study from western Turkey. Energy. advances in flat plate photovoltaic/thermal (PV/T) solar collectors. Renew.
Convers. Manage. 70, 90–106. Sustain. Energy Rev. 15 (1), 352–365.
Azadeh, A., Maghsoudi, A., Sohrabkhani, S., 2009. An integrated artificial neural IEA (International Energy Agency), 2010. Sees Great Potential for Solar, Providing up
networks approach for predicting global radiation. Energy Convers. Manage. 50, to a Quarter of World Electricity by 2050. https://www.iea.org/newsroom/
1497–1505. news/2010/may/2010-05-11–1.html; (accessed 16.09.12).
 A. Asakereh et al. / Solar Energy 155 (2017) 342–353 353

IEA (International Energy Agency), 2013. Energy Policy Highlights. https://www.iea. Ruelland, D., Ardoin-Bardin, S., Billen, G., Servat, E., 2008. Sensitivity of a lumped
org/publications/freepublications/publication/energy-policy-highlights-2013. and semi-distributed hydrological model to several methods of rainfall
html; (accessed 17.4.14). interpolation on a large basin in West Africa. J. Hydrol. 361, 96–117.
Janke, J.R., 2010. Multicriteria GIS modeling of wind and solar farms in Colorado. Saaty, T.L., 1980. The Analytic Hierarchy Process. McGraw-Hill, New York.
Renew. Energy 35, 2228–2234. Saaty, T.L., 2008. Decision Making for Leaders: The Analytic Hierarchy Process for
Kalogirou, S.A., 2004. Solar thermal collectors and applications. Prog. Energy Decisions in a Complex World. RWS Publications, Pittsburgh, Pennsylvania.
Combust. 30, 231–295. Sabziparvar, A.A., 2008. A simple formula for estimating global solar radiation in
Li, D., 2013. Using GIS and Remote Sensing Techniques for Solar Panel Installation central arid deserts of Iran. Renew. Energy 33, 1002–1010.
Site Selection (M.Sc Thesis). University of Waterloo. Sánchez-Lozano, J.M., Teruel-Solano, J., Soto-Elvira, P.L., García-Cascales, M.S., 2013.
Li, D.H.W., Lam, T.N.T., Chan, W.W.H., Mak, A.H.L., 2009. Energy and cost analysis Geographical Information Systems (GIS) and Multi-Criteria Decision Making
of semi-transparent photovoltaic in office buildings. Appl. Energ. 86, 722– (MCDM) methods for the evaluation of solar farms locations: case study in
729. south-eastern Spain. Renew. Sustain. Energy Rev 24, 544–556.
Mahvi, A.H., 2007. Feasibility of solar energy in disinfection of drinking water in Sánchez-Lozano, J.M., Antunes, C.H., García-Cascales, M.S., Dias, L.C., 2014. GIS-
Iran. Am.-Eur. J. Agri. Environ. Sci. 2 (4), 407–410. based photovoltaic solar farms site selection using ELECTRE-TRI: evaluating the
Maleki, S., Firuzi, M.A., Ahmadi, A., 2012. The role of solar energy in sustainable case for Torre Pacheco, Murcia, Southeast of Spain. Renew. Energy 66, 478–494.
development in Khuzestan province, Iran. GJMST 1 (6), 1–6. Shafiee, S., Topal, E., 2009. When will fossil fuel reserves be diminished? Energy
Mirhosseini, M., Sharifi, F., Sedaghat, A., 2011. Assessing the wind energy potential Policy 37 (1), 181–189.
locations in province of Semnan in Iran. Renew. Sustain. Energy Rev. 15, 449– Shao, H., Chu, L., 2008. Resource evaluation of typical energy plants and possible
459. functional zone planning in China. Biomass Bioenergy 32, 283–288.
Moghaddamnia, A., Remesan, R., Hassanpour Kashani, M., Mohammadi, M., Han, D., Singh, J.M., 2002. On Farm Energy Use Pattern in Different Cropping Systems in
Piri, J., 2009. Comparison of LLR, MLP, Elman, NNARX and ANFIS models – with a Haryana, India (MS. Thesis). International Institute of Management, University
case study in solar radiation estimation. J. Atmos. Sol.-Terrest. Phys. 71, 975– of Flensburg.
982. Statistical Centre of Iran, 2011. Khuzestan province. http://www.amar.org.ir
Mondino, E.B., Fabrizio, E., Chiabrando, R., 2014. A GIS tool for the land carrying (accessed 15.03.02).
capaci. Energy Procedia 48, 1576–1585. Sun, Y.W., Hof, A., Wang, R., Liu, J., Lin, Y.J., Yang, D.W., 2013. GIS-based approach for
Mostafaeipour, A., Sedaghatb, A., Dehghan-Niric, A., Kalantarc, V., 2011. Wind potential analysis of solar PV generation at the regional scale: a case study of
energy feasibility study for city of Shahrbabak in Iran. Renew. Sustain. Energy Fujian Province. Energy Policy 58, 248–259.
Rev. 15, 2545–2556. Timilsina, G.R., Kurdgelashvili, L., Narbel, P.A., 2012. Solar energy: markets,
Phuangpornpitak, N., Tia, S., 2011. Feasibility study of wind farms under the Thai economics and policies. Renew. Sustain. Energy Rev. 16, 449–465.
very small scale renewable energy power producer (VSPP) program. Energy Tsoutsos, T., Franztzeskaki, N., Gekas, V., 2005. Environmental impacts from the
Procedia 9, 159–170. solar energy technologies. Energy Policy 33, 289–296.
Pohekar, S.D., Ramachandran, M., 2004. Application of multi-criteria decision Wakeyama, T., Ehara, S., 2010. Renewable energy potential evaluation and analysis
making to sustainable energy planning – a review. Renew. Sustain. Energy Rev. for use by using GIS -a case study of Northern-Tohoku area and Tokyo
8, 365–381. metropolis, Japan. IJESD 1 (5), 446–453.
Rahimikhoob, A., 2010. Estimating global solar radiation using artificial neural Watson, J.J.W., Hudson, M.D., 2015. Regional Scale wind farm and solar farm
network and air temperature data in a semi-arid environment. Renew. Energy suitability assessment using GIS-assisted multi-criteria evaluation. Landscape
35, 2131–2135. Plan 138, 20–31.
Razykov, T.M., Ferekides, C.S., Morel, D., Stefanakos, E., Ullal, H.S., Upadhyaya, H.M., Yue, C., Wang, S., 2007. GIS-based evaluation of multifarious local renewable energy
2011. Solar photovoltaic electricity: current status and future prospects. Sol. sources: a case study of the Chigu area of southwestern Taiwan. Energy Policy
Energy 85 (8), 1580–1608. 35, 383–394.