A Study of The Relationship Between Bioclimatic Comfort Zones And Land Use: The Case of Sivas Province (Turkey)

The aim of this study is to reveal the relationship between bioclimatic comfort zones and land use in Sivas province. In this context, the relationship between the climatic data of 1990 and 2018 and the land use data of Sivas province belonging to the same years was evaluated as seasonal and annual periods. The bioclimatic comfort zones in the study area were determined depending on environmental climatic parameters (ECP) [temperature (T), relative humidity (RH) and wind speed (WS)] and bioclimatic indices [Physiological Equivalent Temperature (PET), Thermo Hygrometric Index (THI), and Universal Thermal Climate Index (UTCI)]. The values of the environmental climate parameters of Sivas province for the relevant years were obtained from 9 meteorological stations, the height of which varies between 1121 m and 1528 m. With the help of the Geographical Information System (GIS), the spatial distribution of the bioclimatic comfort zones determined depending on the environmental climate parameters and bioclimatic indices were created. Land use maps of the study area for reference years were obtained by using CORINE land cover data. The relationship between bioclimatic comfort zones and land use was also determined with the help of GIS. According to the results of this study; It was determined that the land use type in which the bioclimatically comfortable areas overlap in Sivas province differs according to the used parameter/indexes, years and annual periods. a decreasing trend was observed in rural areas. This study aims to reveal the relationship between bioclimatic comfort zones determined based on environmental climate parameters (ECP) and bioclimatic indices (PET, THI, UTCI,) and land use changes in Sivas province. In the study, meteorological data of Sivas province for the years 1990 and 2018 and the CORINE land cover data of the relevant years were used. values, Annual Average Temperature: AAT, Annual Average Relative Humidity: AARH, Annual Average Wind Speed: AAWS The basis of the method applied in this study is to reveal the relationship between the bioclimatic comfort zones, which are determined and mapped depending on the environmental climate parameters and bioclimatic indices of the study area, and the land use in the study area. Details regarding the determination of bioclimatic comfort value ranges depending on ECP, the calculation of bioclimatic index (PET, UTCI and THI) values, and the creation of spatial distribution maps for the relevant parameters and indices are given below. and THI were used to determine suitable areas of Sivas province terms of bioclimatic comfort, depending on bioclimatic indices. order to calculate these monthly average values of temperature (T), relative humidity (RH) and wind speed (WS) parameters were used to obtain and four-season average values of these Bioclimatic comfort areas of Sivas province evaluated In this study, the general climate data of Sivas province (temperature, relative humidity, wind speed) was evaluated and bioclimatic comfort areas of the province in 1990 and 2018 were determined with the help of ECP and PET, UTCI and THI indexes on the basis of these climate data. These comfort areas were evaluated together with the land use in the reference years. In the study, it was determined that the annual average temperature of the province increased by 3 ºC from 1990 to 2018, while the relative humidity and the wind speed decreased by about 10% and 0.3 m/sec, respectively.

RayMan software. Danisvar et al. (2013) evaluated bioclimatic comfort conditions in Iran based on PET method. The study concluded that bioclimatic comfort conditions in the country occur mostly in spring, the researcher emphasized that areas with an altitude between 1000 and 2000 m have better conditions. Ahmadi and Ahmadi (2017) obtained thermal comfort maps by using meteorological parameters of 43 meteorological stations of Iran between 1970-2013. Researchers have demonstrated bioclimatic comfort mapping (BCM) of the study area based on bioclimatic indices such as Temperature Humidity Index (THI), Effective Temperature (ET) and Relative Strain Index (RSI). According to the results of the researchers; Thermal comfort in the northern and western half of Iran is higher than in the southern and eastern parts of the country. The ET and THI results divided the whole country into six regions, ranging from regions with thermal comfort de ciency to regions with thermal comfort conditions. The study concluded that there are no thermal comfort conditions for most of the year in the middle and southeastern regions of the country as well as the southern part of it. In the study by Vinogradova (2020), UTCI was applied to evaluate the bioclimatic situation in Russia. According to the results obtained by the researcher; It has been determined that all cold stress and all heat stress categories are observed in Russia, but cold stress conditions are more dominant. In summer, heat stress and uncomfortable conditions were observed in most of Russia. In most of the country, the maximum UTCI values for the study period corresponded to mild and strong heat stress. Cetin and Sevik (2020) investigated the relationship between land use and bioclimatic comfort zones determined based on PET index in Trabzon province using GIS and prolonging sensing technologies. Bioclimatic comfort maps produced for the years 1985, 1994, 2005 and 2018 and CORINE land use maps belonging to the same years were evaluated together. In the study conducted by Gungor et al. (2021), The bioclimatic comfort areas of Mersin city center between 1972 and 2018 were revealed depending on the PET index. Meteorological parameters such as surface and air temperature, wind speed and relative humidity were taken into account in PET calculations. According to the results of the study; As a city station, PET values increased in areas close to water in Mersin, while a decreasing trend was observed in rural areas.
This study aims to reveal the relationship between bioclimatic comfort zones determined based on environmental climate parameters (ECP) and bioclimatic indices (PET, THI, UTCI,) and land use changes in Sivas province. In the study, meteorological data of Sivas province for the years 1990 and 2018 and the CORINE land cover data of the relevant years were used.

The study area
The province of Sivas is located in Central Anatolia's upper Kzlrmak Region. The province, which has an area of 28.488 km2 and is Turkey's second largest after Konya, is located between 36o and 39o east longitudes and 38o and 41o north latitudes ( Figure 1). Sivas province, in general, is shaped like a plateau, with valleys between single mountains or mountain groups, sunken plains, and hills. The province of Sivas is the coldest in Central Anatolia. Summers are hot and dry, while winters are bone-chillingly cold. Summer is a short period. There are signi cant temperature changes between the summer and winter seasons, as well as between day and night. In the summer, the temperature can reach 40 degrees Celsius, while in the winter, it can plunge to -33 degrees Celsius. The study area generally shows a structure rising towards the north-northeast and south-southeast of the city center. The height of the study area above sea level varies between 581 and 3.012 m (Figure 1).
When the population data of the study area in the reference years are examined; the population of the province in 1990 was 767,481, but it decreased by 15.75% to 646,608 in 2018. While the rural population ratio of the province decreased from 50. [23][24][25][26][27].22% in the relevant years; urban population ratio increased from 49.77-72.78% (TSI, 2021; Table 1). In this study, environmental climate parameters such as temperature (T), relative humidity (RH) and wind speed (WS) for the years 1990 and 2018 of 9 meteorology stations located in Sivas province were used for bioclimatic comfort analysis. Detailed information about the meteorology stations is given in Table 2, and the spatial distribution of the meteorology stations is shown in Figure 1. and the average of September, October and November were used for annual analysis, winter season analysis, spring season analysis, summer season analysis and autumn season analysis, respectively.
In order to determine the relationship between bioclimatic comfort and land use, land use data for the years 1990 and 2018 covering the study area were used.
Land use data for the relevant years of the study area were obtained within the scope of the Coordination of Information on the Environment program (CORINE) coordinated by the European Union (CORINE, 2021).
Excell 2017 software was used for all analyzes of temperature (T), relative humidity (RH) and wind speed (WS) environmental climate parameters (annual and 4 seasons) and Thermo Hygrometric Index (THI) calculations used in the study. Rayman 1.2 software was used for Physiological Equivalent Temperature (PET) calculations (Błazejczyk, 1994), and Bioclima 2.6 software was used for Universal Thermal Climate Index (UTCI) calculations (Błazejczyk, 2017). Within the scope of the CORINE program data, ArcGIS 10.2 software was used to obtain the land use data of the relevant years in the study area.

Method
The basis of the method applied in this study is to reveal the relationship between the bioclimatic comfort zones, which are determined and mapped depending on the environmental climate parameters and bioclimatic indices of the study area, and the land use in the study area. Details regarding the determination of bioclimatic comfort value ranges depending on ECP, the calculation of bioclimatic index (PET, UTCI and THI) values, and the creation of spatial distribution maps for the relevant parameters and indices are given below.

ECP-based BCM analysis
Temperature (T), relative humidity (RH) and wind speed (WS) parameters were used to determine the bioclimatic comfort zones in Sivas province depending on environmental climate parameters. Table 3 shows the environmental climate parameter (ECP) ranges in terms of bioclimatic comfort. Values with a range of 15-20°C in terms of temperature parameter, 30-65% in terms of relative humidity and 0-5 m/s in terms of wind speed parameter were given a score of "1" in the study area and the areas with these values were de ned as "Comfortable" in terms of bioclimatics. The temperature (T), relative humidity (RH) and wind speed (WS) values outside the areas de ned as bioclimatically comfortable were assigned a score of "0" and these areas were also de ned as "Uncomfortable" in terms of bioclimatics (Table 3)

BCM analysis based on bioclimatic indices
In this study, PET, UTCI and THI indices were used to determine suitable areas of Sivas province in terms of bioclimatic comfort, depending on bioclimatic indices. In order to calculate these indices, monthly average values of temperature (T), relative humidity (RH) and wind speed (WS) parameters were used to obtain annual and four-season average values of these indices. Bioclimatic comfort areas of Sivas province were evaluated depending on the annual and four-season average values of these bioclimatic indices. Table 4 and Table 5 show the thermal evaluation ranges of the PET, UTCI and THI indices, respectively. According to Table 4; Evaluation ranges of PET and UTCI indices are divided into 9 and 10 groups, respectively (Zare et al. 2018), and according to Obtaining daily temperature (T), relative humidity (RH) and wind speed (WS) data of meteorology stations, Finding the annual and four-season averages by taking the arithmetic averages of the daily data of the temperature (T), relative humidity (RH) and wind speed (WS) parameters, Lowering the temperature values to sea level with the help of the formula given in Equation (1) below, The next step is to convert the temperature, relative humidity and wind values lowered to sea level with the help of the above equation (1) into PET values with the help of RayMan 1.2 software. For the PET index, the corresponding PET value for the "coldest" thermal class is lower than 4°C, while the corresponding PET value for the "hottest" thermal class is greater than 41°C. PET values in the 18-23°C range represent the "comfortable" class ( . UTCI is accepted as the equivalent temperature for the environment obtained from a reference environment. It is de ned as the air temperature of the reference environment that produces the same strain index value when compared to the response of the reference individual to the real environment. It is one of the most widely used indexes to calculate heat stress in outdoor spaces (Blazejczyk, 1994). UTCI index can be calculated by using dry temperature, average radiation temperature, water vapor pressure or by using dry temperature, relative humidity and wind speed (Zare et al. 2018).
For the UTCI index, the corresponding UTCI value for the "coldest" thermal class is values lower than -40°C, while the corresponding UTCI value for the "hottest" thermal class is values greater than +46°C. UTCI values in the range of 9-26°C represent the "no thermal stress" class (Table 4).  Thom (1959), can be used most widely in bioclimatic studies and is accepted as an indicator of temperature humidity. This index is used in microclimatology studies to evaluate the effect of different temperature and humidity levels on human comfort. THI can be calculated with the help of the formula in Equation (2) below (Ahmadi and Ahmadi, 2017;Emmanuel, 2005).
For the THI index, the corresponding THI values for the "coldest" thermal class are values between -20 °C and -10 °C, while the corresponding THI values for the "hottest" thermal class are values greater than 30 °C. THI values in the 15-20 °C range represent the "comfortable" class (Table 5).

GIS-based BCM analysis
In the study, the Inverse Distance Weighting (IDW) method in the Spatial Analysis module of ArcGIS 10.2 software was used to create the spatial distribution maps of the biolimatic comfort areas depending on environmental climate parameters and bioclimatic indices (PET, UTCI and THI).
Temperature, relative humidity and wind speed parameters, which are effective in determining suitable areas in terms of bioclimatic comfort based on environmental climate parameters, were reclassi ed with the help of ArcGIS 10.8 software according to the suitability scores in terms of bioclimatic comfort (1 or 0) in Table 2. The reclassi ed thematic maps were heavily overlapped by evaluating them together according to the factor weight of each environmental climate parameter with the help of the "Weighted Overlay" module of ArcGIS 10.2 software. Spatial distribution maps showing the suitable areas of Sivas province in terms of bioclimatic comfort were created depending on the environmental climate parameters in line with the weights of these factors.
The calculated values of PET, UTCI and THI indices, which were taken into account in the evaluation of biolimatic comfort zones related to bioclimatic indices, were reclassi ed with the help of ArcGIS 10.8 software according to the evaluation intervals of each index in Table 3 and Table 4. The color levels given in Table 4 and Table 5 for the evaluation intervals of all three indices were used as a reference in showing the spatial distribution of these indices.
The determination of the relationship between bioclimatic comfort zones and land use was performed with the help of the Zonal tool in the Spatial Analyst plugin of ArcGIS 10.8 software.

CORINE land use/cover mapping
The CORINE land use/cover classi cation is designed to include various land covers of European Union (EU) countries. The CORINE land cover classi cation is a classi cation that is standard for the whole of Europe and de nes a land cover of 44 classes arranged hierarchically at three levels. The rst level corresponds to the ve main classes (arti cial areas, agricultural elds, forests and semi-natural areas, wetlands, water bodies). The second level ( In this study; Within the scope of CORINE land use/cover classes, the land use classes of the study area for the years 1990 and 2018 were arranged as 6 classes (natural and semi-natural areas, forested areas, water structures, agricultural areas, arti cial areas and residential areas). In this classi cation, industry-trade-transportation areas, mine-dump-construction areas, non-agricultural green areas were taken into consideration within the scope of "arti cial areas". Maquis and herbaceous plants, open areas that are not covered with vegetation or covered with a small amount of vegetation (bare rocks, sparsely planted areas, etc.) were evaluated within the scope of "natural areas and semi-natural areas". Lake areas such as rivers and dams are taken into consideration within the scope of "water structures". in the summer months. It is understood that there was a decrease in the lowest temperature values and and upward trend in the highest temperature values of the province in the last 28 years ( Figure 2).
Relative humidity: When the relative humidity map of the year 1990 ( Figure 3), which was created on the basis of the relative humidity data of 9 different meteorology stations of the province, is examined; It will be seen that the annual average relative humidity value is between 53.83% and 81.73%. The areas with the highest annual average relative humidity (73.74%-81.73%) are in Ulaş and its surroundings, while the areas with the lowest annual average relative humidity (53.83%-59.63%) are concentrated around Divriği and Gurun. The average relative humidity value of Sivas city center and its surroundings is between 64.22% and 68.49%. In the relevant year, the lowest average relative humidity values in the province were determined in the summer period, while the highest relative humidity values were found in autumn and winter. According

Spatial Distribution of Bioclimatic Comfort Zones
The bioclimatically comfortable areas in Sivas differ according to the environmental climate parameters (ECP) and bioclimatic indices (PET, THI, UTCI) used in the study. In this section, the spatial distribution of bioclimatic comfort areas based on the relevant parameters and indices is interpreted.

Spatial Distribution of Bioclimatic Comfort Zones Based on Environmental Climate Parameters (CIP-ECP)
Considering the analysis made according to the annual average data of the bioclimatic comfort areas based on the environmental climate parameters of the study area, it is seen that comfortable areas increased in 2018 compared to 1990. When the change of comfort areas according to the seasons is examined, it was concluded that the whole of the province is not comfortable in terms of bioclimatics in the winter period in both years. In addition, it was found that comfortable areas increased in spring and autumn, and bioclimatic comfort areas decreased in summer ( Figure 5).

Spatial Distribution of Bioclimatic Comfort Zones Based on Bioclimatic Indexes
Physiological Equivalent Temperature Index (PET) Considering the analysis made according to the annual average data of the bioclimatic comfort areas based on Physiological Equivalent Temperature (PET) Index of the study area, it is seen that cool areas dominate in both years, and in terms of bioclimatic comfort in 1990, the cold areas turned into cool areas in 2018. When the change of comfort areas according to the seasons is examined, it was found that the whole province is in the very cold category in both years and is not comfortable in terms of bioclimatics. The regions that were cold in the spring of 1990 were included in the category of cool regions in 2018. In the autumn season, almost all of the province is in the cool category in 1990, while in 2018, it is seen that the surroundings of Gürün and Divriği moved to the slightly cool category. In the summer months, almost all of the province was found to be suitable in terms of bioclimatic comfort in 1990, while in the same period of 2018, it was concluded that the surroundings of Gurun and Divriği warmed up more than in 1990 ( Figure 6).

Thermo Hygrometric Index (THI)
When the annual map prepared according to the Thermo Hygrometric Index (THI) is evaluated (Figure 7), it is seen that cold and cool areas were dominant in terms of bioclimatic comfort in 1990; In 2018, it was observed that these areas turned into comfortable areas, except for Kangal and its surroundings. When the change of comfort areas according to the seasons is examined, it was determined that the city, which was very cold in the winter of 1990, turned into the cold category in 2018 and was not comfortable in terms of bioclimatics in this period.
While cold regions were dominant in almost all of the province in the spring of 1990, it was determined that most of these areas turned into comfortable areas in 2018. In the summer of 1990, it was determined that these areas, which were hot and very hot areas, turned into comfortable and cool areas in 2018. In the autumn season, it was concluded that almost every region of the province was comfortable except for Kangal and Gurun areas in 1990 and Kangal district in 2018 ( Figure 7).
Universal Thermal Climate Index (UTCI) When the annual map prepared according to the Universal Thermal Climate Index (UTCI) is examined (Figure 8), it is seen that the entire province was under slight cool stress in 1990. However, in 2018 it was observed that some of these areas (regions in the north-eastern and western parts of the province) turned into areas that are not under any thermal stress and are suitable in terms of bioclimatic comfort. When the change of comfort areas according to the seasons is examined, it was determined that while moderately cold stressed areas were dominant in almost all of the province in the winter of 1990, there was a transformation into slightly stressed areas in Sivas city center and Gemerek and Suşehri districts in 2018. In the spring of 1990, areas with slight cold stress spread in almost all of the province, except for Gürün and Kangal. However, it is understood that the bioclimatic comfort classi cation of Sivas city center, Gemerek, Divriği and Suşehri has changed in 2018 and there is no thermal stress in these regions. While there was no thermal stress in the whole province in both years during the summer period, it is observed that there was a change from 1990 to 2018 in the autumn months from areas with slight cold stress to areas without any thermal stress (Figure 8).

Spatial Distribution of Land Use/Cover
When Figure 9 and Table 6 regarding the land use classes of the province in 1990-2018 are examined, it is seen that the land use type that covers the most area in the province in both years is natural-semi-natural areas and agricultural areas. From 1990 to 2018, an increase was observed in natural and seminatural areas, water bodies, arti cial and urban areas, as well as a decrease in forest areas and agricultural areas ( Figure 9, Table 6).

Evaluation of the Relationship Between Bioclimatic Comfort and Land Use
According to the annual climate data of the ECP It was concluded that while the land use type with the highest overlap in terms of bioclimatic comfort in 1990 was natural-semi-natural areas (with a value of 59.85%), only 21.28% of the urban areas in the relevant year were suitable in terms of bioclimatic comfort. In 2018, the land use type where the highest overlap is experienced with areas suitable for bioclimatic comfort was forest areas. From 1990 to 2018, it was determined that the areas suitable for bioclimatic comfort in urban areas increased and reached 73.49% (Table 7-Table 8).
In 1990, the period with the highest bioclimatic comfort was the summer period. In this period, it was found that the rate of areas suitable for bioclimatic comfort in all land use types (forest areas, water bodies, agricultural areas, arti cial areas and urban areas) except natural and semi-natural areas is over 90%.
While 96.30% of urban areas were suitable in terms of bioclimatic comfort in the summer period of 1990, all land use types were not suitable in terms of bioclimatic comfort in the winter period. According to the relevant parameter, it was determined that the highest level of comfortable areas in all land types was spring in 2018, and 89.16% of urban areas were bioclimatically comfortable in the relevant period ( Table 7-Table 8).

According to PET index
It is seen that the period with the highest number of comfortable areas in 1990 was the summer period.  (Table 8).

According to UTCI index
In 1990, it was determined that the period when there was no thermal stress / bioclimatic comfortable areas at the highest rate was the summer period followed by the autumn period. It was found that there is no thermal stress in 100% of all land use types in the summer period, and all areas are bioclimatically comfortable. In the autumn of the same year, it was seen that that 51.59% of natural-semi-natural areas, 52.10% of forest areas, 73.33% of water bodies, 58.97% of agricultural areas, 68.98% of arti cial areas and 76.85% of urban areas are bioclimatic comfortable areas (Table 7).
Looking at the comfort values of the relevant index for 2018, it will be seen that the summer period is again the highest in areas without thermal stress, followed by the autumn period and the spring period. It was determined that in the summer period, 100% of all land use types -as in 1990 -do not have any thermal stress, all areas are comfortable in terms of bioclimatic. In the autumn of the same year, 94.04% of natural-semi-natural areas, 98.21% of forest areas, 87.82% of water bodies, 86.86% of agricultural areas and 94.45% of arti cial areas were bioclimatic comfortable areas. In this period, it was found that the bioclimatic comfort values of urban areas increased by 14.72% compared to 1990. It was determined that 91.57% of the urban areas in the autumn period of the relevant year and 40.96% in the spring period were suitable in terms of bioclimatic comfort (Table 8).

Conclusion And Discussion
Physical planning and design processes are usually conducted with the purpose of creating sustainable and environmentally friendly living spaces, and since these processes concentrate on providing optimal living conditions for people while preserving environmental and cultural elements, the impact of bioclimatic comfort on humans and other species is of particular importance for these processes. In this context, it's critical to assess climatic conditions in a way that leads to a more convenient existence and include them into planning. Similarly, the effect of existing land use and physical features on climate is critical, and it is one of the elements that must be taken into account in sustainable design and planning. Therefore, one of the key tasks in the design of sustainable spaces is to investigate the impacts of climate factors and physical characteristics on each other (Çetin and Şevik 2020).
In this study, the general climate data of Sivas province (temperature, relative humidity, wind speed) was evaluated and bioclimatic comfort areas of the province in 1990 and 2018 were determined with the help of ECP and PET, UTCI and THI indexes on the basis of these climate data. These comfort areas were evaluated together with the land use in the reference years. In the study, it was determined that the annual average temperature of the province increased by 3 ºC from 1990 to 2018, while the relative humidity and the wind speed decreased by about 10% and 0.3 m/sec, respectively.  45 11 It is thought that the most important factor in this situation is urbanization and wrong land use, as is the case all over the world. As a matter of fact, the rate of urbanization in the province in the last 28 years is 23.01%, as the ratio of urban/built up areas and arti cial areas increased by 0.37% in total. On the other hand, it is among the ndings of the study that there is a decrease in agricultural and forest areas. In the study, cold stress conditions are more dominant in the province. However, in the evaluations made according to all parameters and indices, it was revealed that the areas with bioclimatic comfort increased in 28 years, and the density of these areas has shifted from the summer months to the autumn and spring months. It was concluded that the most comfortable areas coincided with the forest and water bodies in both reference years and the parameters/indexes used in the study. This is a proof that the presence of plants and water bodies in cities increases the thermal quality there. When the bioclimatic comfort levels of the urban areas in the relevant reference years were evaluated, it was concluded that the comfortable areas increased in 2018 and their seasonal distributions changed throughout the year. It is possible to say that the climatic comforts of urban areas shift from summer to autumn and spring. By evaluating these data, it is very important to create local strategy plans in urban areas and integrate them into regional plans and policies in order to adapt to climate change studies (Sanchez et al., 2018, Hurlimann et al.,

2021).
The most important result of the study that the bioclimatically comfortable areas (location and areal distribution) in the study area and the land use types where these areas overlap differ according to the parameters/indexes used, the relevant years and different periods of the year. It is actually a normal result that bioclimatic comfortable areas consist differently in each index. Because the parameters evaluated in each index and the comfort ranges and comfort groupings of these parameters are de ned differently. E.g; Temperature, relative humidity and wind speed parameters were evaluated in the ECP used in the study; the study area was evaluated in 2 scales as "Comfortable" or "Uncomfortable" areas in terms of bioclimatics, according to the environmental parameter ranges. This evaluation is a form of evaluation that limits the interpretation of data. Another index used in the study is the PET index. Relative humidity and wind speed parameters are used in PET, which is the most widely used index in bioclimatic comfort analysis. Daneshvar  Sivas province. The most important fundamental difference of the study from the related studies is that more than one parameter and indices are used in determining the suitable areas in terms of bioclimatic comfort, and therefore it can reveal the differences in approach between the indices. The second important difference is that the bioclimatically comfortable zones change annually and seasonally (in 5 different periods) in 2 different 28-year time periods (1990 and 2018) according to each parameter/index. The third important difference is that the climatically comfortable zones in the reference years and the land cover of the province are evaluated together.
As a result, this study is a base study for the planning of sustainable urban models that are more compatible with the environment, people and nature in this   Mapping and spatial distribution of land use/cover based on CORINE in 1990 and 2018