The methodology of the study comprises eight steps as follows: 1) literature review and data collection of the study area and solar power plants; 2) determination of analytical instruments such as GIS, MCDM, FAHP; 3) environmental analysis including impact assessment systems principles and legal framework; 4) excluding the restricted areas while gathering together the data for technical analyses; 5) determination of the layers used in technical analyses based on different features suitable with impact assessment systems principles as well; 6) application of the analytical instruments 7) identification of the sensitive areas by technical and environmental analyses, and 8) achieving outcomes, outputs and proposals (Fig. 3).
3.2 Determination of Criteria
In order to provide the highest level of results in the studies carried out for SPP site selection and to make an accurate analysis, the data, which is the most basic need, must be provided in an appropriate way (Memişoğlu, 2014; Çolak et. al., 2020; Shorabeh et. al, 2022; Goh et. al, 2022; Karipoğlu et. al, 2022). In determining the areas suitable for SPP installation, layers representing the study area were selected as a result of both field and literature studies (Asakereh et al., 2017; Uyan, 2013; Noorollahi et al., 2016; Al Garni and Awasthi, 2017). These layers are the factors of solar radiation, distance to PDCs, slope, land use, distance to fault line, geology, distance to stream, distance to road line, aspect and distance to residential areas. Since Safranbolu district is not on a migratory bird route (URL-1, 2022), the distance to migration routes parameter was not used in the analysis.
The design and performance analysis of solar energy systems depend on solar radiation data (Bulut and Büyükalaca, 2007; Kaygusuz and Ayhan, 1999). The most important criterion for the efficiency of solar energy systems is the solar radiation affecting that region. Solar radiation directly affects the energy to be obtained from the power plant to be installed. Solar radiation is a fundamental input in renewable energy applications (Nourani et al., 2019). For this reason, the solar radiation parameter has been examined in many studies (Charabi and Gastli, 2011; Effat, 2013; Kengpol et al., 2013; Sánchez-Lozano et al., 2013; Uyan, 2013). Therefore, solar radiation data was used as a parameter in this study and the solar radiation map (MENRT, 2022) is given in Fig. 5a.
The distance to the substation centre is economically very important in determining the optimal installation areas for SPP. The reason for this is that it is more costly to transmit the electricity generated as it moves away from the centre. It is preferred that the areas where SPP will be installed are located less than 10 km from substations (Uyan, 2017). The distance factor to the substation in the study area was evaluated under 5 classes (Fig. 5b).
Topography of the land is among the most important factors in determining the optimal installation areas for SPPs. Slope and aspect, which are among the topographic factors, have a direct effect on the selection. It is desired to have a low slope where SPP panels will be installed. In general, areas with a slope above 11% are considered unsuitable, while slopes of 4% and less than 4% are considered quite suitable. One of the reasons for this is that solar panels shade each other when the slope is high. This has a negative effect on efficiency. The other is that the high slope increases the installation cost of the power plant (Noorollahi et al., 2016). Despite all these, on the other hand, completely flat lands without slopes are not preferred for SPP installation due to water accumulation and accumulated water drainage problems (Şenlik, 2017). The slope data of the study area was produced from the digital elevation model of the region. The slope map produced is divided into 5 classes (Fig. 5c).
Regarding the aspect, the slope of the land in the south direction has a significant effect on the duration and amount of insolation and increases the suitability of the land (Effat, 2013; Noorollahi et al., 2016). For this reason, regarding SPP site selection, it is necessary to prefer the southern facades that provide the most benefit from the sun during the day and provide high sun angle according to seasonal conditions (Yalçın and Yüce, 2020; Sarsıcı, 2020). The aspect map of the study area was evaluated to include a total of 9 classes in the range of -1 to 360 degrees (Fig. 5d).
Existing land uses are very effective for determining the optimal locations for SPP. Settlement areas, military areas, areas with protection status (National Park, Nature Park, Wildlife Development Area, etc.), transport networks, agricultural lands, forests, etc. within the administrative boundaries of the area where the research is carried out should be determined as a priority, and the physical and legal limitations of the study should be determined accordingly. SPPs should be at a certain distance from these regions according to the characteristics of these regions. In addition, it should not be ignored that settlements grow in parallel with the population growth and even SPPs may stay within these regions. Since the SPP will affect the natural life, cultural and natural protection areas in the environment where it is established, it should be at a certain distance from the protection areas. Areas such as forests, maquis, reeds, etc. are definitely not preferred in the selection of the location, since increasing the shading of natural vegetation will reduce the efficiency of electricity generation (Şenlik, 2017). In this study, land use was analysed in 5 classes under the headings of agriculture, forest, pasture, mixed vegetation and settlement (Köseoğlu, 2015) and the land use map is shown in Fig. 5e.
During the construction process (installation) of the SPP plant, being close to transport networks is an important criterion that ensures that both the cost of infrastructure installation and the maintenance and repair costs of the plant are low. In addition, utilising the existing transport routes will prevent the opening of new roads for transport purposes and will prevent the increase of different land uses in the immediate vicinity and cultural pressures on environmental components (Yücel, 2016; Uyan, 2017). However, SPPs should be more than 100 m away from the roads so that solar panels are not affected by the pollution caused by the traffic on the roads (Uzar and Koca, 2020). Taking all these into consideration, the distance factor to the road lines in the study area was evaluated under 6 classes (Fig. 5f).
Hydrological structure is another effective factor in determining the optimal areas for SPP. When considering the criterion of proximity to the river, it is desired that SPPs should not be closer than 400 m to the rivers, taking into account all the changes that rivers and river beds will show in different seasons (Uzar and Koca, 2020). There are 4 canyons in the study area, named Tokatlı, Düzce (Kirpe), Sırçalı and Sakaralan (Yaci). Since the locations of these canyons are within the restricted area in terms of proximity to rivers, they are considered within the distance to rivers parameter and not discussed separately. In the study area, the distance to the river factor was evaluated under 5 classes (Fig. 5g).
Geological structure is another parameter used in determining SPP installation location alternatives (Uyan, 2017) and affects the cost of the power plant project (Eroğlu, 2018). In addition, it is desirable that the ground where the power plant to be installed should present a geological structure that will allow the application of the hammer mounting system. There are 12 different lithological formations and 4 members surfacing the study area. Stratigraphically, the formations in the study area are Aydos (Oa), Ereğli (ODe), Ferizli (Df), Yılanlı (DCy), Bürnük (Jb), Ulus (Ku), Safranbolu (Tes), Karabük (Teka), Soğanlı (Teso), Akçapınar (Tea), Örencik (Tplö) formations and alluvium (Qal). Aydos Formation (Oa) consists of quartzitic sandstone and conglomerate and is 50–200 metres thick. Eregli formation (ODe), which consists of shale, sandstone and limestone, has a thickness of 300–550 metres. Ferizli Formation (Df) consists of dolomite, sandstone, algea and oolitic ironstone and has a thickness of about 100 metres. The Yılanlı formation (DCy), which consists of limestone, dolomitic limestone and dolomite successions, is 1000 metres thick. Bürnük formation (Jb) consists of conglomerate, sandstone, carbonate sandstone and reaches a thickness of 200 metres. Ulus formation (Ku), which consists of shale, claystone, marl, limestone, sandstone, sandy limestone and conglomerate succession, is divided into 2 members, Ahmetusta (Kua) and Sunduk (Kus). The conglomerates within the Ulus Formation constitute the Ahmetusta Member (Kua), and the rocks are grey, yellow, brownish yellow, medium-thick layered and massive in appearance. The Sunduk Member (Kus) consists of grey, beige, medium-thick layered limestones. The thickness of the formation can reach up to approximately 2000 metres. The Safranbolu Formation (Tes) starts with a thin conglomerate-sandstone level at the bottom and transitions to carbonate sandstone, sandy limestone and limestone level towards the top and its thickness varies between 50–500 metres. The Karabük Formation (Teka), which has a medium-thick layered characteristic, consists of marl at the lower levels and claystone and sandstone succession towards the top. There are also thin coal levels in the upper parts of the formation. The Çerçen Member (Tekaç), which is a member of the Karabük Formation, consists of conglomerate, sandstone, siltstone, claystone and mudstone succession and is about 450 metres thick. The Soğanlı Formation (Teso) consists of limestones and there are marl layers between the limestone layers. There are abundant joints and deep karst structures within the limestones of the unit with an average thickness of 150 metres. Akçapınar Formation (Tea) consists of dolomitic limestone, claystone, mudstone and marl succession, mainly clayey limestone. The formation has a thickness of 150–200 metres. The Örencik Formation (Tplö), the younger deposits of the study area, consists of terrestrial conglomerate, sandstone and mudstone succession. It shows medium-thick stratification and offers a thickness of 50–100 meters. Yörük Member (Tplöy), who is a member of this formation, consists of lacustrine limestones and has a thickness of approximately 100 meters. The youngest unit of the study area is Quaternary aged alluvium (Qal) and consists of river beds, old pits and gravel, sand and mud sediments developing on flat areas (Timur and Aksay, 2002; Gedik and Aksay, 2002).
Since the alluvium in the study area consists of uncemented sediments, it is not a suitable layer for installing SPP on it. However, if it is necessary to establish an SPP on the alluvium, it will be possible to implement it provided that the hammered mounting system is used and the system used is lowered until reaching to the foundation under the alluvium. Otherwise, it does not offer a secure feature for SPP installation. All geological units in the study area offer suitable features for the hammer mounting system. The Safranbolu Formation (Tes) and the Sunduk Member (Kus), which is a member of the Ulus Formation, contain deep and widespread karst structures such as canyons and caves, therefore, detailed field work is required for SPP installation on them as they offer dissolible property. In addition, SPP should never be installed in karstic environments that are part of tourism activities and contain natural geological heritage elements that need to be protected. The geology map of the study area is given in Fig. 5h.
In the study, seismicity was also examined within the scope of the geological structure. As the proximity to the active fault lines increases the earthquake effect, in site selection analyses, the distance to fault lines is also a very effective factor. Areas with less earthquake risk are more suitable for SPP installation. For this reason, the fault line data was obtained by digitizing from the Fault Map presented on the website of the General Directorate of Mineral Research and Exploration, MTA (MTA, 2022; AFAD, 2022) and five different buffer areas were created at 2500 m intervals to be used in the analysis (Fig. 5i).
Transmission of energy generated by SPP to remote consumption regions causes energy loss. For this reason, in order to avoid loss of efficiency in energy transmission, remote locations to residential and industrial areas should be avoided in SPP site selection (Şenlik, 2017), provided that SPPs do not remain within the expanding residential areas over time (Uzar and Koca, 2020). Considering all these, the distance factor to the residential areas was evaluated under 5 classes in this study (Fig. 5j).
Considering the data of current situation given in Fig. 5 in terms of environmental-social and economic-based impact assessment principles for SPP establishment,
(a) Safranbolu district is advantageous in terms of renewable energy potential since the annual total solar radiation data is in the range of 1400–1550 KWh/m2/year.
(b) In case of selecting SPP installation areas that are relatively closer to the substation (less than 10 km), there is no need to establish new transmission networks that require a costly process. By integrating the regions with these conditions into the technical process; it is aimed to control the occurrence of cultural pressures to a certain extent and to prevent additional negative impacts on economic, social and natural basis.
(c) By integrating areas with a slope ≤ 4% into the technical analysis process, it is aimed to prevent the generation of excavation waste to a certain extent by supporting minimal excavation-filling works. In such a situation, the change in the integrity of the natural soil system will be kept under control to a certain extent.
(d) Due to the fact that Türkiye is located in the northern hemisphere, the fact that the panels are positioned predominantly in the south direction in SPP systems increases the efficiency. This requirement has been integrated into the technical analysis process.
(e) Among the land uses, there are dense flora such as trees, shrubs, vegetation cover, meadow-pasture, etc. where the presence of flora is intense. Care has been taken to exclude such areas from the selection criteria in order to avoid destruction of ecological integrity.
(f) The activity is expected to have an easily accessible road, railway, airway and/or maritime road network in order to provide the transport required especially during the construction process. The fact that these roads are among the existing ones is not only economically preferable, but also a matter that considers the integrity of the natural system. For this reason, areas in the range of 100–500 m have been integrated into the technical analysis process. In addition, the connections to be made should not damage the existing village roads and should be planned in a way that will not prevent the traffic flow and endanger the life and property safety of the people.
(g) The fact that the facility area selection is close to the underground-ground water resources is a situation that poses a risk to the environment and public health. Within the scope of the Regulation on the Protection of Potable Water-Utility Water Basins published and enacted in the Official Gazette dated 28.10.2017 and numbered 30224, it is recommended to select it from the areas located at least 400 m away. This threshold was taken in to account in the study. In the works to be carried out, necessary precautions should be taken by considering the water resources, no waste should be left in the water resources and no intervention should be made and no waste water should be discharged to the underground.
(h) In the study area, it is observed that geological and geomorphological formations with unique characteristics remain within restricted areas.
(i) During the operation period in the project area, the site selections made among the areas far from active fault lines in accordance with the " Türkiye Building Earthquake Regulation" published in the Official Gazette dated 18.03.2018 and numbered 30364 will provide economic and environmentally based use in the long term.
(j) The effects of the noise and wastes to be generated due to the activity during the installation and operation phase on the nearest settlements is an issue that should be taken into consideration in accordance with the "Regulation on the Assessment and Management of Environmental Noise" published in the Official Gazette dated 04.06.2010 and numbered 27601. For this reason, attention has been paid to the location of the areas integrated into the technical analysis at a distance from the dense residential areas.
3.3. Method
3.3.1. Fuzzy Logic and Fuzzy Decision Making
In cases where a decision needs to be made with incomplete or uncertain information, fuzzy logic methods are preferred. It is possible to define decision-making as the determination of the opinions that must be finalised at all levels, by addressing all aspects of the problem, sometimes as a single issue and sometimes as a series of issues, aiming to obtain the most accurate result (Harcar, 1992). The presence of verbal information in decision-making processes may cause uncertainty and this may lead to the creation of models with subjective thoughts. In other words, if the uncertainties cannot be eliminated, it is recommended to analyze by accepting the existence of uncertainty. Performing analytical solutions with fuzzy logic provides the decision maker with a more flexible decision environment. Using fuzzy logic, it is possible to obtain the numerical equivalents of the verbally presented information and put them into the solution process. If personal data such as questionnarie results and expert opinions are included in the solution process, the solutions may differ according to the number of the respondent group and the change, which affects the optimal decision (Aydın, 2009).
Fuzzy logic is based on rating. Since there are fixed and precise rules in the measurability of variables, Zadeh (1965) proposed fuzzy models as an alternative approach. Fuzzy logic, which has been increasingly important since 1965, is defined as a mathematical order established to explain uncertainties and work with uncertainties (Klir and Yuan, 1995).
3.3.2. Fuzzy Multi-Criteria Decision Making (FMCDM) and Fuzzy Analytic Hierarchy Process (FAHP)
MCDM models can be seen in which decision makers are not objective in their judgments, express their judgments verbally, or their judgments do not contain precise and complete information. In such cases, analyzes in decision models can be made with a fuzzy logic approach.
Applications of fuzzy logic in the decision-making process are usually carried out by blurring classical decision theories. In decision problems defined by fuzzy logic, just like in classical problems, it is aimed to reach the non-fuzzy "best" decision. However, fuzzy theory aims to reveal in which probability each alternative can be optimal rather than the optimal decision reached. In other words, it is important to apply the methods developed with fuzzy theory in cases where there are no definite determinations in the problems, the parameters are not known precisely or the evaluations are verbal (Klir and Yuan, 1995).
AHP is a MCDM method based on binary comparisons and comparisons can be made subjectively or objectively depending on the definition of the criteria. It is necessary to determine the comparison weights objectively, considering how important one alternative is compared to the other. On the other hand, when comparisons are made according to the criteria of expert opinions such as suitability, preferability and importance, where personal evaluations come to the fore, subjective results arise. Although the subjectivity of expert opinions is seen as an advantage of the AHP method (Anderson et al., 1997), this personalisation makes the results less precise. For this reason, FAHP, which was first proposed in the work of Laarhoven and Pedrycz (1983), is preferred in such cases. In this study, judgements and weights are expressed in triangular fuzzy numbers. Buckley (1985), on the other hand, argued that a single solution cannot always be obtained in the work of Laarhoven and Pedrycz (1983) and worked with trapezoidal fuzzy numbers based on arithmetic operations of fuzzy numbers. Different from these, Lee et al. (1999) introduced the concept of interval for binary comparisons based on probabilistic optimisation in fuzzy comparisons. Chang (1996) improved the method by using synthetic degrees. The first step in FAHP is to express the problem in a hierarchical structure showing the objective, criteria, sub-criteria and alternatives (Awasthi et al., 2018). In the second stage, a numerical link is established between the objective and the criteria. In this study, geometric mean method (Buckley, 1985) and the Extended Analysis Method (Chang, 1996) have been used from fuzzy analytic hierarchical process methods. In Table 1 the triangular fuzzy scale implemented in the current study has been introduced. After the fuzzy synthesis values are calculated, these values are compared with each other and the priority values of the options and criteria are found. By normalising this vector, the real priority vector is obtained.
Table 1
The scale of fuzzy AHP pair-wise comparison (Felix et al., 2008).
| Linguistic Variables | Fuzzy scale | Reciprocal fuzzy |
| Equally important | (1,1,1) | (1, 1, 1) |
| More important | (2/3, 1, 3/2) | (2/3, 1, 3/2) |
| Much more important | (3/2, 2, 5/2) | (2/5, 1/2, 2/3) |
| Too important | (5/2, 3, 7/2) | (2/7, 1/3, 2/5) |
| Definitely important | (7/2, 4, 9/2) | (2/9, 1/4, 2/7) |
3.3.3. Environmental Impact Assessment (EIA) Process
Implementation of impact assessment systems for sectoral activities with spatial dimension is important in terms of ensuring integrated environmental management. In this framework, when the direct or indirect relationship of the energy sector with other sectors and different land uses is considered. It is clear that decision-making processes need to be subjected to an impact assessment process. An EIA process focuses on the location and technical conditions of the planned project. Within the scope of this study, among the issues considered while conducting the EIA process is primarily the determination of suitable locations for SPP. Therefore, in determining the criteria in the technical analysis process carried out within the scope of the study, the principles of impact assessment systems have also been taken into consideration. The main reason for this is the fact that the areas determined to be suitable for installation and their immediate surroundings will be within the scope of the impact area during the installation and operation of the project.
In the process of determining the areas to be proposed for the purpose of SPP facility, EIA principles observe that the most suitable areas in terms of environmental, social and economic aspects are determined and this situation is presented with a cause-effect relationship. In this direction, considering the solar energy data of Safranbolu district, electricity production can be supported by SPP systems; considering the population density, it has been determined that a systematic production process will support an efficient supply at the city level. Detailed and complete implementation of the legal framework for SPP facilities, of which location proposals have been developed with the aim of establishing and operating in the city, prevents long-term negative effects on the environment. The establishment of SPP in Safranbolu will ensure that the rich natural resources of the city are utilised in accordance with their purpose. As a result, natural resources will be utilised as input to the national economy.
At the end of the technical analysis process, location alternatives compatible with EIA principles were determined. The results of the analysis were analysed in comparison with the F29 map section of the 1/100,000 scale Environmental Plan of the Zonguldak-Bartın-Karabük Planning Region and it was investigated whether there were any legal contradictions (Directorate General of Spatial Planning, 2022). In addition, it has been investigated whether the optimal areas determined within the scope of the study are areas that need to be protected by taking into account the Sensitive Areas in Annex 5 of the EIA Regulation.
(i) "National Parks", "Nature Parks", "Nature Monuments" and "Nature Conservation Areas" defined in Article 2 of the National Parks Law No. 2873 and designated in accordance with Article 3 of this Law
In the determination of "National Parks", "Natural Parks", "Natural Monuments" and "Nature Conservation Areas" within the scope of Safranbolu district, institutional interviews were conducted and the Wildlife Database of the General Directorate of Nature Conservation and National Parks has been scanned. There are no areas defined as "National Parks", "Natural Parks", "Natural Monuments" and "Nature Conservation Areas" in the area subject to the study and its immediate vicinity and there are no areas with legal status in this context.
(ii) "Wildlife Protection Areas, Wildlife Development Areas and Wild Animal Settlement Areas" determined in accordance with the Land Hunting Law No. 4915
In the determination of "Wildlife Protection Areas and Wild Animal Settlement Areas" within the scope of Safranbolu district, institutional interviews were conducted and the Wildlife Database of the General Directorate of Nature Conservation and National Parks has been scanned. Accordingly, it was determined that "Sırçalı Wildlife Development Area" is located within the administrative boundaries of Safranbolu. The areas whose coordinates were determined were accepted as restricted areas.
(iii) In the 1st, 2nd, 3rd and 5th subparagraphs of the fırst paragraph titled "Definitions" of Article 3 of the Law No. 2863 on the Protection of Cultural and Natural Assets, the areas defined as "Cultural Assets", "Natural Assets", "Sites" and "Protected Areas" and the areas identified and registered in accordance with the relevant articles of the same Law and the Law dated 17/6/1987 and numbered 3386 (on the Amendment of Certain Articles of the Law No. 2863 on the Protection of Cultural and Natural Assets and the Addition of Certain Articles to this Law)
Within the scope of the literature review and institutional interviews conducted for Safranbolu district; the coordinate information of the areas with special status defined as "Cultural Assets", "Natural Assets", "Protected Areas" and "Protected Areas" identified within the borders of the district were determined, and these areas were accepted as restricted areas although these areas have been determined as optimal areas. Within the scope of the environmental analysis carried out within the scope of the study, different legal bases were examined. When the determined optimal areas are evaluated in terms of international conventions;
(i) Bern Convention: Hazard categories of existing plant species IUCN Hazard Categories (Red Data Book of Turkish Plants).
It is seen that there are no plant species in the IUCN categories among the plant species detected in the determined areas and their immediate surroundings. None of the plant species in the study area are included in the Bern Convention Annex-1 list.
(ii) Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES).
There is no plant species included in the CITES Convention among the plant species detected in the determined areas and their immediate surroundings. The CITES Convention is a contract that binds the import and export of wild animal and plant species, in short, international trade with certain permits and documents, between the contracting countries.
(iii) Convention on the Protection of Wetlands of International Importance, Especially as Waterfowl Habitat (RAMSAR Convention).
There are no areas protected by the Ramsar Convention in the class of "Class A Wetlands" according to international criteria in the designated areas and their immediate surroundings.
There is no restrictive situation within the scope of other international conventions to which Türkiye is a party (Espoo Convention, Barcelona Convention, European Landscape Convention, etc.).
Relevant legal limitations in the Turkish legislative system: In this context, on the base of the "Principle Decision on Solar Power Plant (SPP) in Natural Protected Areas" published in the Official Gazette dated 25.01.2017 and numbered 29959, which aims to establish SPP areas by considering the integrity of the natural system and on the base of the "Principle Decision numbered 100. amending the Resolution on Solar Power Plants (SPP) in Natural Protected Areas" published in the Official Gazette dated 20.07.2022 and numbered 31898, Qualified Natural Protected Areas and Sustainable Conservation and Controlled Use Areas within the administrative boundaries of Safranbolu district are accepted as restricted areas in the process of determining the optimal area for SPP (SAYS, 2022). After determining the location for the SPP facilities planned to be established in Türkiye, in accordance with the EIA Regulation, which is included in the Turkish legislative system (given in the Introduction Section) and entered into force after being published in the Official Gazette dated 29.07.2022 and numbered 31907, the relevant provisions of the Regulation are guiding the legal process according to the capacity of the planned activity.
3.4 Implementation
It is also necessary to determine a method for the selection of criteria to be used in the MCDM method to be used in solving problems in the field of renewable energy (Wang et al., 2009). The general characteristic of decision-making problems is fuzziness and FAHP allows the preferences of decision makers to be expressed in adaptable ways (Güler and Yomralıoğlu, 2018). In addition, classical MCDM methods cannot reveal the ambiguity of human thought too much (Ecer, 2018). For all these reasons, the FAHP method was used in this study. Firstly, the fuzzy number equivalents of the linguistically expressed criteria determined by the experts were determined and the binary comparisons of the criteria obtained in line with the expert opinions are shown in the binary comparison matrix in Table 2.
Table 2
Pairwise comparison matrix for FAHP (a solar radiation b distance to PDCs, c slope, d aspect, e land use, f distance to road line, g distance to stream, h geology, ı distance to fault line, j distance to residential areas)
| | a | b | c | d | e | f | g | h | ı | j |
| A | (1,1,1) | (2/3,1,3/2) | (3/2,2,5/2) | (3/2,2,5/2) | (5/2,3,7/2) | (5/2,3,7/2) | (7/2,4,9/2) | (7/2,4,9/2) | (7/2,4,9/2) | (7/2,4,9/2) |
| B | (2/3,1,3/2) | (1,1,1) | (2/3,1,3/2) | (3/2,2,5/2) | (3/2,2,5/2) | (5/2,3,7/2) | (5/2,3,7/2) | (5/2,3,7/2) | (7/2,4,9/2) | (7/2,4,9/2) |
| C | (2/5,1/2,2/3) | (2/3,1,3/2) | (1,1,1) | (2/3,1,3/2) | (3/2,2,5/2) | (3/2,2,5/2) | (5/2,3,7/2) | (5/2,3,7/2) | (7/2,4,9/2) | (7/2,4,9/2) |
| D | (2/5,1/2,2/3) | (2/5,1/2,2/3) | (2/3,1,3/2) | (1,1,1) | (2/3,1,3/2) | (2/3,1,3/2) | (3/2,2,5/2) | (3/2,2,5/2) | (5/2,3,7/2) | (5/2,3,7/2) |
| E | (2/7,1/3,2/5) | (2/5,1/2,2/3) | (2/5,1/2,2/3) | (2/3,1,3/2) | (1,1,1) | (2/3,1,3/2) | (2/3,1,3/2) | (2/3,1,3/2) | (3/2,2,5/2) | (3/2,2,5/2) |
| F | (2/7,1/3,2/5) | (2/7,1/3,2/5) | (2/5,1/2,2/3) | (2/3,1,3/2) | (2/3,1,3/2) | (1,1,1) | (2/3,1,3/2) | (2/3,1,3/2) | (3/2,2,5/2) | (3/2,2,5/2) |
| G | (2/9,1/4,2/7) | (2/7,1/3,2/5) | (2/7,1/3,2/5) | (2/5,1/2,2/3) | (2/3,1,3/2) | (2/3,1,3/2) | (1,1,1) | (1,1,1) | (2/3,1,3/2) | (2/3,1,3/2) |
| H | (2/9,1/4,2/7) | (2/7,1/3,2/5) | (2/7,1/3,2/5) | (2/5,1/2,2/3) | (2/3,1,3/2) | (2/3,1,3/2) | (1,1,1) | (1,1,1) | (2/3,1,3/2) | (2/3,1,3/2) |
| I | (2/9,1/4,2/7) | (2/9,1/4,2/7) | (2/9,1/4,2/7) | (2/7,1/3,2/5) | (2/5,1/2,2/3) | (2/5,1/2,2/3) | (2/3,1,3/2) | (2/3,1,3/2) | (1,1,1) | (1,1,1) |
| J | (2/9,1/4,2/7) | (2/9,1/4,2/7) | (2/9,1/4,2/7) | (2/7,1/3,2/5) | (2/5,1/2,2/3) | (2/5,1/2,2/3) | (2/3,1,3/2) | (2/3,1,3/2) | (1,1,1) | (1,1,1) |
For the extent analysis method that will be used we need to calculate the value of fuzzy synthetic extent with respect to the \({i}\) parameter (Chang, 1996) shown in Eq. 1,
$${S}_{i}=\sum _{j=1}^{3}{M}_{{g}_{i}}^{j}⨀{\left[\sum _{i=1}^{7}\sum _{j=1}^{3}{M}_{{g}_{i}}^{j}\right]}^{-1}$$
1
where all the M_(g_i)^j, j = 1,2,3 are triangular fuzzy numbers lying in Table 2.
$${S}_{1}=\left(23.67 28 32.5\right)⨀\left(\frac{1}{167.38},\frac{1}{136.50},\frac{1}{110.32}\right)=\left(0.141, 0.205, 0.295\right)$$
The fuzzy synthetic extent values and normalized weights for the parameters are given in Table 3.
Table 3
The fuzzy synthetic extent values and weights for the parameters according to the Chang method
| \({i}\) | Parameter | Synthetic extent values (\({\tilde{{S}}}_{{i}}\)) | Weights (\({{w}}_{{i}}\)) |
| 1 | Solar Radiation | 0.141, 0.205, 0.295 | 36.82 |
| 2 | PDC | 0.118, 0.176, 0.258 | 29.45 |
| 3 | Slope | 0.106, 0.161, 0.233 | 24.89 |
| 4 | Aspect | 0.070, 0.110, 0.171 | 8.84 |
| 5 | Land use | 0.046, 0.076, 0.124 | 0.00 |
| 6 | Distance to road | 0.045, 0.075, 0.122 | 0.00 |
| 7 | Distance to stream | 0.035, 0.054, 0.088 | 0.00 |
| 8 | Geology | 0.035, 0.054, 0.088 | 0.00 |
| 9 | Distance to fault | 0.030, 0.044, 0.069 | 0.00 |
| 10 | Distance to residential areas | 0.030, 0.044, 0.069 | 0.00 |
Using the geometric mean method proposed by Buckley (1985), the fuzzy weights and weights for the parameters are given in Table 4.
Table 4
The fuzzy weights and weights for the parameters according to the Buckley method
| \({i}\) | Parameter | Fuzzy weights (\({\tilde{{w}}}_{{i}}\)) | Weights (\({{w}}_{the determined optimal areas {i}}\)) |
| 1 | Solar Radiation | 0.14, 0.21, 0.31 | 20.18 |
| 2 | PDC | 0.12, 0.18, 0.27 | 17.43 |
| 3 | Slope | 0.10, 0.15, 0.23 | 14.68 |
| 4 | Aspect | 0.07, 0.12, 0.19 | 11.93 |
| 5 | Land use | 0.05, 0.08, 0.13 | 8.26 |
| 6 | Distance to road | 0.04, 0.07, 0.12 | 7.34 |
| 7 | Distance to stream | 0.04, 0.05, 0.09 | 5.50 |
| 8 | Geology | 0.04, 0.05, 0.09 | 5.50 |
| 9 | Distance to fault | 0.03, 0.04, 0.07 | 4.59 |
| 10 | Distance to residential areas | 0.03, 0.04, 0.07 | 4.59 |
As a result of FAHP weighting with different sorting methods, it was calculated that solar radiation criterion has the highest weight, followed by distance to substation, slope, aspect, land use, distance to road, distance to stream, geology, distance to fault lines and distance to settlement centres. Considering the calculated weights of the parameters, the maps in Fig. 3 were combined and the restricted areas were removed from this map and SPP suitability maps were created for the study area.