Environ Sci Pollut Res https://doi.org/10.1007/s11356-017-0424-1

RESEARCH ARTICLE

Thermal map assessment under climate and land use changes; a case study for Uzundere Basin Nalan Demircioglu Yıldız 1 & Ugur Avdan 2 & Sevgi Yılmaz 1 & Andreas Matzarakis 3

Received: 17 February 2017 / Accepted: 5 October 2017 # Springer-Verlag GmbH Germany 2017

Abstract Satellite images are commonly used for defining thermal urban environment and heat islands in urban areas. The objective of this study was to analyze and evaluate thermal environment of Uzundere-Erzurum using satellite images provided by the Landsat Thematic Mapper (TM) sensor on board Landsat 5 satellite. In respect to land use and urban surface features, spatial characterization of thermal urban environment was defined. In order to define the link between surface emissivity, land surface temperatures (LST), and urban surface characteristics more effectively, the CORINE Land Cover (CLC) database for Uzundere was also used. Based on the Landsat satellite images, different land cover categories are defined, and the alterations of heat islands are identified according to these categories. The minimum and maximum surface temperatures were obtained from open spaces, non-vegetated bare rocky areas (− 1.8 and 29.1 °C, respectively). The surface temperature in the water areas was much better (7.5 and 21.9 °C) and followed by green areas surface temperature (− 0.9 and 28 °C). Surface temperatures should be considered energy-focused planning.

Responsible editor: Philippe Garrigues * Sevgi Yılmaz [email protected]; [email protected]

1

Architecture and Design Faculty, Department of Landscape Architecture, Atatürk University, 25240 Erzurum, Turkey

2

Research Institute of Earth and Space Sciences, Anadolu University, Eskişehir, Turkey

3

Research Centre Human Biometeorology, Deutscher Wetterdienst, Stefan-Meier-Str. 4, D-79104, Freiburg, Germany

Keywords Thermal band . Landsat . Land use . Uzundere-Erzurum-Turkey . Land surface temperature

Introduction Changes in climate and land-use type threat to environment globally (Stathopoulou and Cartalis 2007; Cai 2001). Although the human effects on land surface varies widely, almost every part of the land surface more or less affected by human activities (Weng et al. 2004). Land is used for producing food, fiber, energy, biochemicals, and medicines, and also as a place to live. Therefore, land use has always changed with changes in human needs because of changes in cultural, socioeconomical, and environmental conditions (Ostwald and Chen 2006; Rounsevell and Reay 2009; Zhao et al. 2011). Regional development and environmental changes are strongly dependent on land use and climate. In the last decades, many scientists focused on analyzing effects of climate and land-use changes on environment. Yan et al. (2013) emphasized that assessing effects of climate and land-use changes is essential for sustainable development and a key factor for human reaction to global change. The size, population, and industrial development of a city, topographical structures (Kohl 1999; Hais and Kucera 2009), physical layout, regional climate (Kang and Cressie 2013), and meteorological conditions (Beltrami 2001; Luhar et al. 2006; Yilmaz et al. 2016) affect the intensity of the temperature differences between urban and rural sites. Extensive urbanized surfaces alter the energy and water balance processes and affect the dynamics of air movement (Oke 1987). Because of human activities, urban heat islands (UHI) are considerably warmer than that of its surrounding rural areas. The studies on the characteristics of the UHI have been well-documented. Deosthali (2000) pointed out that the core of the city appeared as both heat and moisture islands at night, but it was heat and dry

Environ Sci Pollut Res

islands at the time of sunrise. The weakest average maximum UHI occurs in summer but strong in fall and winter (Zhong 1996; Kim and Baik 2002). It was found that the UHI intensity was inversely correlated with the rural temperature; however, the spatial extent was independent from heat island magnitude and rural temperature (Streutker 2002; Chen et al. 2006). The thermal environmental patterns of extensive urban areas can be effectively depicted from satellite thermal data. A number of studies on surface UHI have been carried out using thermal data from satellites (Aniello et al. 1995; Streutker 2002; Lo and Quattrochi 2003; Dousset and Gourmelon 2003; Avdan et al. 2014). Identification of urban areas using thermal infrared data acquired by satellite has first time been demostrated by Rao (1972). Medium scale sensors have been used in these studies (Landsat TM, NOAA AVHRR) (Gallo and Owen 1998), which give a spatially continuous view of the surface UHI over large urban areas. Meteorological station network data might be used for recognizing the correlation between land surface temperature and land cover using land cover maps extracted from satellite data (SPOT HRV, Landsat TM, Landsat MSS, Landsat 7 ETM). Landsat Thematic Mapper (TM) and Enhanced Thematic Mapper Plus (ETM+) thermal infrared (TIR) data high spatial resolutions have also been used for small-scale studies of UHI (Chen et al. 2002 ). It was shown by the Research on Land Surface Temperature (LST) studies that surface soil water content and vegetation cover significantly affect on surface radiant temperature response (Owen et al. 1998), which also encouraged studies on defining relationship between LST and vegetation abundance (Gallo and Owen 1998). Weng et al. (2004) emphasized that applying height measurements and elevation data to the UHI studies are important, because topography and morphology both have an effect on the measurement of fractal dimensionality from thermal surfaces.The main goal of this study was to analyze and evaluate the thermal environment of land-use changes using satellite images in Erzurum/Uzundere in Turkey. Uzundere-watershed was chosen, as the study area is inhabited by many animal and plant species and is recognised as the natural beauty museum for not only the region but also for Turkey. The area, having an important place in the life of people in its proximity and contributing to the economy of both region and the country, has also significance for the conservation of both ecosystem and biodiversity. In the Uzundere-watershed management, the aim is to conserve the natural reserves with a sustainable point of view and with the participation of the people at decision-making process by contributing to the socio-cultural and economic development of the native people in the area. The processing method on computing LST in ERDAS IMAGINE and the models used to estimate effects of land-use changes in urban and rural environments have been presented in this paper. It was believed that the results of this study may help to urban planners and managers for developing new strategies on urban design and drawing landscaping policies for reducing the negative effects of climatic changes.

Materials and methods Landsat satellite data used in this study were provided from the Mineral Research & Exploration General Directorate (MTA) archive. The data were pre-processed and projected to the Universal Transverse Mercator (UTM) WGS 84 projection system. The Uzundere district is located in the north of Erzurum, Turkey, (N 40° 31′ 48″ E 41° 32′ 23″) (Fig. 1). The Landsat satellite image of the study area for the year of 1987 (16-08-1987 (cloud cover 0%, Path/Row: 172/32) is shown in the Fig. 1. The boundary of the study area is cropped from Landsat Satellite data (RGB/4 3 2). The reason why this Landsat image (1987) was chosen is that Provincial Directorate of Agriculture owns this image which is convenient to CORINE land use type (Anonymous 1978). Cloudless images were preferred from USGS Earth Explorer website (http://glovis.usgs.gov/) when choosing the satellite images taken in 1987s. The temperature data of Erzurum and its surrounding towns for the accuracy analysis of the temperature maps were provided by the General Directorate of Meteorology of Turkey. The LSTs were derived from the Landsat TM Thermal Infrared band (10.45–12.42 μm). Satellites were used to estimate surface temperature and the atmospheric correction to radiation from the Earth’s surface (Price 1984). In order to convert the digital number (DN) of Landsat TM TIR band into spectral radiance (Lλ ), the following Eq. (1) was used (Chander et al. 2007; Barsi et al. 2007).  Lλ ¼

 LMAX λ −LMIN λ ðQcal −Qcalmin Þ þ LMIN λ Qcalmax −Qcalmin

ð1Þ

or Lλ ¼ Grescale  Qcal þ Brescale where: 

 LMAX λ −LMIN λ  Qcalmax −Qcalmin  LMAX λ −LMIN λ Qcalmin ¼ LMIN λ − Qcalmax −Qcalmin Grescale ¼

Brescale where: Lλ Qcal Qcalmin Qcalmax LMINλ

Spectralradianceatthesensor’saperture(W/(m2 srμm)) Quantized calibrated pixel value (DN) Minimum quantized calibrated pixel value corresponding to LMINλ (DN = 0) Maximum quantized calibrated pixel value corresponding to LMAXλ (DN = 255) Maximum at-sensor spectral radiance that is scaled to Qcalmin (W/(m2 sr μm))

Environ Sci Pollut Res

Fig. 1 Landsat TM satellite image (RGB/4 3 2) of the study area, green area (1), open area (2), water area (3)

LMAXλ Grescale

Minimum at-sensor spectral radiance that is scaled to Qcalmax (W/(m2 sr μm)) Band-specific rescaling gain factor ((W/(m2 sr μm))/DN)

Brescale

Band-specific rescaling bias factor (W/(m2 sr μm))

Table 1 gives the same values of constant (Chander and Markham 2003; NASA 2004; Chander et al. 2007).

Environ Sci Pollut Res Table 1 Landsat TM 5 spectral range and calibration dynamic range

Satellite/ sensor

Band no.

Spectral range (μm)

Center wavelength (μm)

LMIN (W/(m2 sr μm)

LMAX (W/(m2 sr μm)

Grescale (W/(m2 sr μm)/ (DN)

Brescale (W/(m2 sr μm)

Landsat 5/TM

Band 6

10.45–12.42

11.435

1.2378

153,032

0.055376

1.18

In the second step, the spectral radiance was converted into the sensor brightness temperature using Eq. (2) (NASA 2004; Chander and Markham 2003): K2 lnð1 þ K 1 =Lλ Þ

TB ¼ where TB K1

Effective at-sensor brightness temperature (K) Calibration constant 1 (W/(m2 sr μm))

Fig. 2 NDVI image of the study area

ð2Þ

K2 Lλ ln

Calibration constant 2 (K) Spectral radiance at the sensor’s aperture (W/(m2 sr μm)) Natural algorithm

K1 and K2 are the calibration constants for the Landsat TM 5 (K1 = 607.76 (m2 sr μm; K2 = 1260.56 K) (Chander and Markham 2003). The temperature values were obtained and referenced to a black body. Therefore, corrections for spectral emissivity (ε) were necessary according to the nature of land

Environ Sci Pollut Res Table 2

Calculation of emissivity using NDVI values

NDVI

Land surface emissivity (ε)

NDVI ≤ 0.185

0.995

− 0.185 ≤ NDVI < 0.157

0.970

0.157 ≤ NDVI ≤ 0.727 NDVI > 0.727

1.0094 + 0.047ln(NDVI) 0.990

cover (Weng et al. 2004; Li et al. 2012; Yildiz et al. 2013; Yilmaz et al. 2015). Emissivity of an object is the ratio of energy radiated by an object at a given temperature. The emissivity of natural surface can vary significantly in vegetation characteristics (NASA 2004). The emissivity can be estimated by using Normalized Difference Vegetation Index (NDVI) (Zhang et al. 2006; Giannini et al. 2015). NDVI is one of the most widely used indexes in

Fig. 3 Emissivity image of the study area

monitoring vegetation cover (Kumar et al. 2012; Liu and Zhang 2011; Giannini et al. 2015). The basic algebraic structure of a spectral index comes from the ratio between two spectral bands Red and Near Infrared (NIR). This index is calculated by subtracting Red reflectance from NIR reflectance, and dividing by sum of the two as given in Eq. (3) (McFeeters 1996; Wan and Dozier 1996).

NDVI ¼

NIR−red NIR þ red

ð3Þ

The NDVI image of the Uzundere boundary is shown in Fig. 2. Vandegriend and Owe (1993) defined NDVI values diversity from 0.157 to 0.727 in Eq. (4). A complete land surface emissivity can be calculated using NDVI in Table 2 (Zhang et al. 2006; Liu and Zhang 2011).

Environ Sci Pollut Res Table 3 Stations used for controlling the temperature map

Station name

North

East

Altitude

Near-surface temperature (°C)

Calculated LST (°C)

Error (°C)

İspir Tortum

40,4860 40,3016

40,9995 41,5400

1217 1581

22.0 20.1

21.0 21.0

0.9 − 0.9

Erzurum center Oltu

39,9059

41,2542

1869

20.1

23.2

− 3.1

40,5494

41,9955

1296

20.0

20.5

− 0.5

The emissivity image of the Uzundere boundary is shown in Fig. 3. εi ¼ 1:0094 þ 0:047lnðNDVI Þ

ð4Þ

After the calculation of brightness temperature (TB) and emissivity (ε) images, the final LST image was computed using Eq. (5) (Artis and Carnahan 1982). T¼

TB 1 þ ðλ  T B =∝Þ  lnε

where

T TB λ α h c K ε

Surface radiant temperature (K), The blackbody temperature (K), Wavelength of emitted radiance (λ = 11.435 μm) (Chander et al. 2009a, 2009b) hc/K (1.438 × 10−2 mK) Planck’s constant (6.26 × 10−34 J/s), Velocity of light (2.998 × 108 m/s), Boltzman constant (1.38 × 10−23 J/K), surface emissivity

ð5Þ The near-surface air temperature station was used to verify the final LST (Table 3).

Fig. 4 Landuse map of the study basin (according to Provincial Directorate of Agriculture)

Environ Sci Pollut Res

Fig. 5 Topographic map of study area

This study presents a methodology for estimating the surface UHI intensity of land using Landsat 5 thermal images of high spatial resolution (120 m) in combination with urban land cover information provided by the CORINE Land Cover (CLC) database in a GIS framework. The CLC is a geographic land cover/land use database for a pan-European region. It is produced by the European Environment Agency and provides consistent information on land cover across Europe using a nomenclature of 44 standard classes which are organized in three levels of detail (European Commission 1993). The CLC database for Uzundere in vector format is used for effectively correlating land cover and surface emissivity. For this purpose, the original 44 classes of CORINE were grouped into only 3 ones: Bopen area (1),^ Bgreen area (2),^ and Bwater area (3)^; thus, a new land cover polygon map was produced.

Results A land-use map was prepared using the data obtained from the Provincial Directorate of Agriculture, which use landuse scale

of Corine when determining land use classes. In the study basin, the areas such as meadows, dry and irrigated cultivated lands, non-vegetated mountainous, bare land, and fallow lands where vegetation cover is not continuous are accepted to be open areas (1), forests, shrubs and garden areas as green areas (2), and lake areas and water surfaces are shown in blue (3) while cream color demonstrates mountainous open areas (Fig. 4). In this study, LST is analyzed on non-vegetated mountainous open areas (1), vegetated woody and shrubbery areas (2), and water surfaces on valley plain (3) in Uzundere Basin. Surface temperatures of different land uses show different values between 29.1 and − 1.8 °C. Topographic map (DEM) of study area is given in Fig. 5. According to this map, while the lowest altitude is 788 m, the highest altitude is 3061 m. There is 2273 m altitude difference between highest and lowest ones. Temperature map of study area which is produced through thermal band analysis is given in Fig. 6. When the LST and land use maps are overlapped, the open areas in the application field cover 27,524.06 ha, the forest areas cover 16,919.83 ha and water surfaces cover 655.99 ha. While the

Environ Sci Pollut Res

Fig. 6 LST image of the study area

maximum temperature in open areas was to 29.1 °C, it was 21.9 °C in the water surface areas. In the basin, the open areas mean temperature is 15.7 °C, the temperature in forest areas is 12.4 °C, and the temperature in waters areas is 10.4 °C. As the maximum temperature is measured in non-vegetated rocky (29.1 °C) or soil covered open areas, minimum temperature is measured as − 1.8 °C in these areas. It is measured on green areas and water surfaces respectively as − 0.9 and 7.5 °C. Although, average and maximum values of temperature on water surfaces are measured as low, minimum temperature is 7.5 °C as the highest value. This can be derived from the late heating and cooling characteristics of water surfaces (Table 4). Overlay of different land-use map and temperature map is given in Fig. 7. While water surfaces and green areas are generating coolness starting from − 1.8 °C temperature, there Table 4 Mean temperatures in the study area

are hot places having 29.1 °C temperature value in nonvegetated bare rocky areas. Figure 7 produced according to different land uses and directions in the study area shows temperature decrease from green to yellow color, and temperature increase toward red color. It shows temperature changes from different point of view as green and red colors. While east side of the basin is generally hot and showed with red color, west side is observed as cold and displayed with green color. It can be said according to the results of analysis that vegetated places are observed as cooler than the other areas, whereas bare rocky areas are detected warmer. As it is seen in the Fig. 4, settlement areas and immediate surroundings are observed as green. However, heat stress is clearly observed as high in bare lands. Morphological change or mobility affects vegetation as it affects heat.

Land use

Area (ha)

Min (°C)

Max (°C)

Range

Mean

STD

Sum

Open areas (1) Green areas (2) Water areas (3)

27,524.06 16,919.83 655.99

− 1.8 − 0.9 7.5

29.1 28.0 21.9

31.0 28.7 14.4

15.7 12.4 10.4

5.2 4.7 1.5

5,785,850.0 2,323,174.3 75,666.6

Environ Sci Pollut Res

Fig. 7 Land use (open areas, green areas, water areas) and LST maps overlay

Table 4 represents the estimated surface temperature values over different land uses. Maximum surface temperature is depicted in open areas (1) ranging from − 1.8 to 29.1 °C having a mean surface temperature of 15.7 °C and standard deviation of 5.2 over green areas in green areas (2) temperature difference ranges from − 0.9 to 28.0 °C, having a mean surface temperature 12.4 °C and a standard deviation of 4.7 while over water surface (3) temperature ranges from 7.5 to 21.9 °C, having a mean surface temperature of 10.4 °C and standard deviation of 1.5. The lowest temperature was observed in open areas (− 1.8 °C), followed by green areas (− 0.9 °C) and water surface (7.5 °C). Based on the differences in the characteristics of water surfaces such rivers, ponds, and waterfall, their surface temperatures may vary and cause high standard deviation (1.5 for the present study).

Discussion As emphasized by Yuksel and Yilmaz (2008), the alteration in land uses creates distinctive changes in local, regional, and even global climate by changing the surface roughness,

topography, and the loop between water vapor and greenhouse gas in atmosphere and land surface. By the rise of urbanization, air temperature rises, the winds get weaker and mist, cloudiness, the amount of rain and solar radiation are altered. Matson et al. (1978) defined that the difference between urban and rural areas ranges from 2.6 to 6.5 °C in terms of mean temperatures. It is seen that the results of the present study can show similarity with the results of previous ones. It was found from different studies given in Table 5 that the coolest area which shows the lowest temperature change rate is water surface. It was also found in the present study that the highest temperature difference is in open space (range 31.0; Table 5), while the lowest one is in water surface (range 14.4; Table 5). Water surface was determined to be the coolest. It can be seen when the table is taken into consideration that water surface is the coolest, while bare lands are the warmest. According to the study (Yildiz et al. 2014), while the maximum temperature in dry farming area raised up to 25.1 °C, it was 22.6 °C in the open areas. In the basin, the irrigated area mean temperature is 19.5 °C, the temperature in meadow area is 18.6 °C, and

Environ Sci Pollut Res Table 5

Results of thermal band analysis of land uses in different years

Bare land (open) mean °C

Forest green mean °C

Water surface mean °C

Date

Sensor

References

39.2

29.2

26.5

22.10.1999

Landsat-7 ETM+

Mallick et al. 2008

27.3

24.6

19.2

15.08.2005

Landsat ETM+

Li et al. 2012

24.1

22.3

18.1

11.04.1997

Landsat TM

Li et al. 2009

33.4

31.6

27.1

19.07.2004

Landsat TM

Li et al. 2009

39.6

27.8

22.9

25.10.2009

Landsat TM

Mallick et al. 2012

44.9

32.9

22.7

17.09.1990

Landsat-7 ETM+

Falahatkar et al. 2011

48.9

34.1

38.0

7.09.2001

Landsat-7 ETM+

Falahatkar et al. 2011

34.6

25.1

22.6

16.01.1989

Landsat TM

Van and Bao 2008

36.2 36.6

26.5 29.1

26.9 30.1

25.01.1998 25.12.2006

Landsat TM ASTER

Van and Bao 2008 Van and Bao 2008

the temperature in settlement area is 20.7 °C. As the maximum temperature is measured in non-vegetated open areas (31.7 °C), minimum temperature is measured as 4.8 °C in settlement areas. The coldest areas are the water surface areas and the surrounding. Yuksel and Yilmaz (2008) reported that the temperature in surfaces covered by asphalt and concrete is high, it is low in green areas, and the difference in air temperatures between these areas is up to 15 °C. A similar result was also reported by Yuksel and Yilmaz (2008) that the heat difference between asphalt and grass was 11.8 °C. Xu et al. (2010) found that along the riverside of Huangpu River, the places which are 10–20 m from the river’s edge are believed to have the appropriate values for thermal comfort. Hawkins et al. (2004) found that in the month of April the temperature difference in agricultural areas are set to 0.8–5.4 °C. Grasslands and fruit gardens are defined as cool places, raw, and hard soils, and also the fallow lands are defined as warm places. A project made by Upmanis et al. (1998) also tried to understand the effects of green areas to their surroundings by conducting studies in parks. In a field of 1500 m, these green areas reduce air temperature by 3 °C. Also, Gomez et al. (1998) found out that in Valencia (Spain), the green areas have the same effect by the number of 2.5 °C. Spranken and Oke (1998) in Vancouver defined that these effects are even increased by the thickness of green areas up to 5 °C. This study will be useful to making better decisions about land use planning. In the fields of application, presenting how the urban microclimate gets affected is important for a healthy and comfortable urbanization. Especially in recent years, the urban renewal projects and defining the suitable land for new urbanization areas reveal the importance of this study more. The local governments need accurate and up-to-date data to make the best decisions about land use planning. This kind of

studies should be used in this decision-making process. Humanity needs more comfortable and more livable places. Therefore, the need for urban and rural planning stages is an undeniable fact that the use of reliable data is vital for sustainable and ecological-based planning.

Conclusion Urban growth rate reached an unexpected size in many places around the world. The rise of urbanization and increase urban population reveals the necessity to pay attention on urban climate and urban planning. Temperature is an important meteorological factor in the process of forming the climate. The urban development and expansion lead to an increase of LST over the urban areas. This study was conducted to analyze the spatial distribution of land surface temperature in terms of land use types using Landsat7 TM+. The results showed that the LST has significantly affected by different urban development intensities. Concerning the relation between land use and LST, it is found that evergreen forest and deep water have the lowest LST, while open areas have the highest LST. It is found that LST degrees vary conversely depending on water or moisture content, biomass, and elevation of each land use type.Vegetation and environmental conditions such as vegetation biomass and soil moisture content greatly characterize the relationship between LST and NDVI. Therefore, characterization of surface temperature with land use and cover types may help the decision maker for better future planning. The LST studies using satellite thermal infrared data suggest that different land cover types have distinctive responses. As a result of land conversion from natural and vegetated surfaces into urbanization may result in rising temperature and increasing the spatial variability of LST. The results of this study indicated that the methodology used is quite appropriate to analyze climatic and land use change effects. Of course, the quality of the

Environ Sci Pollut Res

results strongly depends on the quality of data provided. By using the Landsat satellite images of the application area, different land cover categories are defined, and according to these categories, the alteration in the heat islands is found. This study will contribute to the urbanization studies related to the locational preferences. Measuring the effects of preferred land uses on urban microclimate has crucial importance for the healthy and comfortable urbanization. Importance of the study becomes very clear when the recent urban regeneration projects and locational preferences of new settlements are considered. All in all, it can be said that climate must be considered as an important factor for the location of new settlements in order to provide advantage of solar energy to these settlements. This kind of study will be useful to making better decisions about land use planning.

References Aniello C, Morgan K, Busbey A, Newland L (1995) Mapping microurban heat islands using Landsat TM and a GIS. Comput Geosci 21(8):965–969 Anonymous (1978) Turkey land ownership. General Directorate of Soil Survey and Soil Mapping Head of Department 1978 Artis DA, Carnahan WH (1982) Survey of emissivity variability in thermography of urban areas. Remote Sens Environ 12(4):313–329 Avdan U, Demircioglu Yildiz N, Dagliyar A, Yigit Avdan Z, Yilmaz S (2014) Thermal band analysis of agricultural land use and its effects on bioclimatic comfort: the case of Pasinler. Geophys Res Abstr 16: EGU2014–15277 Barsi JA, Markham BL, Helder DL and Chander G (2007) Radiometric calibration status of Landsat-7 and Landsat-5. Proceedings of SPIE 6744, Florence, Italy, Sept. 17–20,URL:http://landsat.gsfc.nasa.gov/ pdf_archive/publications/SPIE07_LandsatRadiometry.pdf Beltrami H (2001) On the relationship between ground temperature histories and meteorological records: a report on the Pomquet station. Glob Planet Chang 29(3–4):327–348 Cai YL (2001) A study on land use/cover change. Geogr Res 20:645–652 Chander G, Markham B (2003) Revised Landsat-5 TM radiometric calibration procedures and postcalibration dynamic ranges. IEEE Trans Geosci Remote Sens 41(11):2674–2677 Chander G, Markham BL, Barsi JA (2007) Revised Landsat-5 thematic mapper radiometric calibration. IEEE Geosci Remote Sens Lett 4(3):490–494 Chander G, Markham BL, Helder DL (2009a) Summary of current radiometric calibration coefficients for Landsat MSS, TM, ETM+, and EO-1 ALI sensors. Remote Sens Environ 113(5):893–903 Chander G, Markham BL, Helder DL (2009b) Summary of current radiometric calibration coefficients for Landsat MSS, TM, ETM+, and EO-1 ALI sensors. Remote Sens Environ 113(5):893–903 Chen Y, Wang J, Li X (2002) A study on urban thermal field in summer based on satellite remote sensing. Remote Sensing for Land and Resources 4:55–59 Chen XL, Zhao HM, Li PX, Yin ZY (2006) Remote sensing image-based analysis of the relationship between urban heat island and land use/ cover changes. Remote Sens Environ 104(2):133–146 Deosthali V (2000) Impact of rapid urban growth on heat and moisture islands in Pune City, India. Atmos Environ 34:2745–2754

Dousset B, Gourmelon F (2003) Satellite multi-sensor data analysis of urban surface temperatures and landcover. ISPRS J Photogramm Remote Sens 58(1–2):43–54 European Commission (1993) CORINE land cover; guide technique, Report EUR 12585EN. Office for Official Publications of the E uropean Communities, Luxembourg Falahatkar S, Hosseini SM, Soffianian AR (2011) The relationship between land cover changes and spatial-temporal dynamics of land surface temperature. Indian J Sci Techno 4(2):76–81 Gallo KP, Owen TW (1998) Assessment of urban heat island: a multisensor perspective for the Dallas-Ft. worth, USA region. Geocarto Int 13:35–41 Giannini MB, Belfiore OR, Parente C, Santamaria R (2015) Land Surface Temperature from Landsat 5 TM images: comparison of different methods using airborne thermal data. J Eng Sci Technol Rev 8(3):83–90 Gomez F, Gaja E, Reig A (1998) Vegetation and climatic changes in a city. Ecol Eng 10(4):355–366 Hais M, Kučera T (2009) The influence of topography on the forest surface temperature retrieved from Landsat TM, ETM and ASTER thermal channels. ISPRS J Photogramm Remote Sens 64(6):585–591 Hawkins TW, Brazel AJ, Stefanov WL, Bigler W, Saffell EM (2004) The role of rural variability in urban heat island determination for Phoenix, Arizona. J Appl Meteorol 43:476–486 Kang EL, Cressie N (2013) Bayesian hierarchical ANOVA of regional climate-change projections from NARCCAP Phase II. Int J Appl Earth Obs Geoinf 22:3–15 Kim YH, Baik JJ (2002) Maximum urban heat island intensity in Seoul. J Appl Meteorol 41:651–653 Kohl T (1999) Transient thermal effects below complex topographies. Tectonophysics 306(3–4):311–324 Kumar KS, Bhaskar PU, Padmakumari K (2012) Estimation of land surface temperature to study urban heat island effect using LANDSAT Etm+ image. Int J Eng Sci 4(2):771–778 Li JJ, Wang X, Wang X, Ma W, Zhang H (2009) Remote sensing evaluation of urban heat island and its spatial pattern of the Shanghai metropolitan area, China. Ecol Complex 6:413–420 Li YY, Zhang H, Kainz W (2012) Monitoring patterns of urban heat islands of the fast-growing Shanghai metropolis, China: using time-series of Landsat TM/ETM+ data. Int J Appl Earth Obs Geoinf 19:127–138 Liu L, Zhang YZ (2011) Urban Heat Island analysis using the Landsat TM data and ASTER data: a case study in Hong Kong. Remote Sens 3(7):1535–1552 Lo CP, Quattrochi DA (2003) Land-use and land-cover change, urban heat island phenomenon, and health implications: a remote sensing approach. Photogramm Eng Remote Sens 69(9):1053–1063 Luhar AK, Venkatram V, Mi LS (2006) On relationships between urban and rural near-surface meteorology for diffusion applications. Atmos Environ 40(34):6541–6553 Mallick J, Yogesh K, Bharath BD (2008) Estimation of land surface temperature over Delhi using Landsat-7 ETM+. J Ind Geophys Union 12(3):131–140 Mallick J, Singh CK, Shashtri S, Rahman A, Mukherjee S (2012) Land surface emissivity retrieval based on moisture index from LANDSAT TM satellite data over heterogeneous surfaces of Delhi city. Int J Appl Earth Obs Geoinf 19:348–358 Matson M, McClain EP, McGinnis DF, Pritchard JA (1978) Satellite detection of urban heat islands. Mon Wea Rev 106:1725–1734 McFeeters SK (1996) The use of the Normalized Difference Water Index (NDWI) in the delineation of open water features. Int J Remote Sens 17(7):1425–1432 NASA 2004. Landsat project science office: Landsat 7. Science Data Users Handbook, Chapter 11. http://landsathandbook.gsfc.nasa.gov/pdfs/ Landsat7_Handbook.pdf Oke TR (1987) Boundary layer climates. Methuen, New York

Environ Sci Pollut Res Ostwald M, Chen D (2006) Land-use change: impacts of climate variations and policies among small-scale farmers in the Loess Plateau, China. Land Use Policy 23(4):361–371 Owen TW, Carlson TN, Gillies RR (1998) An assessment of satellite remotely-sensed land cover parameters in quantitatively describing the climatic effect of urbanization. Int J Remote Sens 19:1663–1681 Price JC (1984) Land surface temperature measurements from the split window channels of the. NOAA-7 AVHRR, J Geophys Res: Atmospheres (1984–2012) 89(D5):7231–7237 Rao PK (1972) Remote sensing of urban Bheat islands^ from an environmental satellite. Bull Am Meteorol Soc 53:647–648 Rounsevell MDA, Reay RS (2009) Land use and climate change in the UK. Land Use Policy 26(1):160–169 Spranken SRA, Oke TR (1998) The thermal regime of urban parks in two cities with different summer climates. Int J Remote Sens 18(11): 2085–2104 Stathopoulou M, Cartalis C (2007) Daytime urban heat islands from Landsat ETM+ and Corine land cover data: an application to major cities in Greece. Sol Energy 81:358–368 Streutker DR (2002) A remote sensing study of the urban heat island of Houston, Texas. Int J Remote Sens 23(13):2595–2608 Upmanis H, Eliasson I, Lindqvist S (1998) The influence of green areas on nocturnal temperatures in a high latitude city. Int J Climatol 18: 681–700 Van TT, Bao HDX (2008) A study on urban development through landsurface temperature by using remote sensing: in case of Ho Chi Minh City. VNU J Sci, Earth Sci 24:160–167 Vandegriend AA, Owe M (1993) On the relationship between thermal emissivity and the normalized difference vegetation index for natural surfaces. Int J Remote Sens 14(6):1119–1131 Wan Z, Dozier J (1996) A generalized split-window algorithm for retrieving land-surface temperature from space, Geoscience and Remote Sensing. IEEE Transactions 34(4):892–905 Weng QH, Lu DS, Schubring J (2004) Estimation of land surface temperature-vegetation abundance relationship for urban heat island studies. Remote Sens Environ 89(4):467–483

Xu J, Wei Q, Huang X, Zhu X, Li G (2010) Evaluation of human thermal comfort near urban waterbody during summer. Build Environ 45(4): 1072–1080 Yan D, Schneider UA, Schmid E, Huang HQ, Pan L, Dilly O (2013) Interactions between land use change, regional development, and climate change in the Poyang Lake district from 1985 to 2035. Agric Syst 119:10–21 Yildiz ND, Avdan U, Yilmaz S (2013) Analyze of climate features with thermal band in Çoruh River’s Valley Base. In: 6th atmospheric science symposium – ATMOS, 24–26 April. Oral Presentation, ITU, İstanbul Yildiz ND., Avdan U., Yilmaz S., Dagliyar A. and Matzarakis A., 2014. Thermal Band Analysis of Different Land Uses in Urban Spaces and its Effects. Third International Conference on Countermeasures to Urban Heat Island, October 13–15, 2014, pp. 547–557, Oral Presentation, Venezia, Italy Yilmaz S, Avdan U, Yilmaz H, Yildiz ND, Irmak MA (2015) Quantification of thermal bioclimate of Erzurum based on different land uses and thermal band information. ICUC9- 9th International Conference on Urban Climate jointly with 12th Symposium on the Urban Environment, 20–24 July. Oral Presentation, Toulouse, France Yilmaz S, Koç A, Mutlu E, Yildız DN (2016) integration of thermal comfort information with spatial modelling in erzurum city center. 4th International Conference on Countermeasures to Urban Heat Island (UHI) 2016. Procedia Eng 169:80–87 Yuksel U, Yilmaz O (2008) A study on determining and evaluating summertime urban heat İslands in Ankara at regional and local scale utilizing remote sensing and meteorological data. Gazi Univ Eng Arch Fac 23(4):937–952 Zhang JQ, Wang YP, Li Y (2006) A C++ program for retrieving land surface temperature from the data of Landsat TM/ETM+ band6. Comput Geosci 32(10):1796–1805 Zhao C, Fu G, Liu X, Fu F (2011) Urban planning indicators, morphology and climate indicators: a case study for a north–south transect of Beijing, China. Build Environ 46:1174–1183 Zhong B (1996) Urban heat island effect of Shenzhen city. J Meteorol 22(5):23–24