2.1. Study Area
Ankara, one of the metropolitan cities located in the Central Anatolia Region of Turkey, was chosen as the main material of the study. The city of Ankara is located at 39° 55' north latitude and 32° 50' east longitude and covers an area of approximately 25,632 km². Since the urban heat island is concentrated especially in city centers and built-up areas, the boundaries of the research area in Ankara were determined by considering this factor. In this context, an area of approximately 1565,558 km² covering the city center and its surroundings was determined as the research area. Settlement areas in Ankara city have a spatial pattern extending east-west. There are climate differences from place to place in the city. There is a steppe climate in the south and a temperate and rainy climate in the north. The hottest month in the city is July-August, and the coldest month is January. The geographical location of the study area is shown in Fig. 1.
Figure 1. Maps of the study area
2.2. Methodology
In the study, the effect of impermeable surfaces, green areas, and water surfaces on LST in Ankara city center between 2013 and 2023 is questioned. LST and LULC indices maps for 2013, 2018, and 2023 were made using Landsat satellite image bands. The results obtained were then evaluated with statistical analysis and graphics (Fig. 2).
Figure 2. Workflow of the study
2.3 Remote Sensing Data
LULC indices are used to determine the urbanization rate and urban surface characteristics. Landsat 8 satellite images with 30*30 resolution provided by the United States Geological Survey (USGS) were used to map the LST and LULC indices. Satellite images for the years 2013, 2018, and 2023 were obtained for temporal and spatial determination of LST and LULC indices. The climate situation of Ankara city was effective in determining the dates of satellite images. In this context, satellite images were selected considering the month of July (Table 1).
Table 1
Description of the satellite data.
Satelies Scene
|
Spatial Resolution
|
Path-Row
|
Date
|
Landsat 8 OLI/TIRS C2 L2
|
30*30
|
P:177
R: 032/033
P:178
R:032
|
16/07/2013
25/07/2013
|
23/07/2018
30/07/2018
|
17/07/2023
28/07/2023
|
2.4 Land Surface Temperature (LST)
ArcGIS 10.4.1 program and the 10th band of the Landsat 8 satellite image were used to create the LST map. The steps followed in making the map are as follows (Yuan and Bauer 2006; Barsi et al., 2014; Roy et al. 2020):
First, Eq. 1 was used to convert the data of the 10th band of the Landsat 8 OLI satellite image into spectral radiance values:
where L λ : TOA spectral radiance (Watt/(m2× srad × m)), ML: Radiance Mult Band (Band10), QCal: quantized and calibrated standard product pixel values: Band10, AL: Radiance Add Band (10). All of this information is obtained from the meta-data file of satellite images.
Then, temperature values were calculated from the brightness values of the images. Eq. 2 was used for the calculation:
T= \(\frac{K2}{In(\frac{K1 }{L\lambda }+1)}-273.15\) (2)
where T: Temperature degree (in Celsius), K 2 : K2 Constant Band(10), K1: K1 Constant Band(10), Lλ: Spectral Radiance. All of this information is obtained from the meta-data file of satellite images.
In the third step, after calculating the temperature degree, a Normalized Vegetation Difference Index (NDVI) analysis was performed. Bands 4 and 5 of the Landsat 8 satellite image were used in NDVI analyses. Eq. 3 was used in the calculation:
NDVI= \(\frac{(Band5-Band4)}{(Band5+Band4)}\) (3)
Formüllerde; Band 4 ve 5 uydu görüntüsü bandlarını temsil etmektedir.
NDVI analizini yapıldıktan sonra, bitki örtüsü oranı hesaplanmaktadır. Bu değer, NDVI analizi kullanılarak yapılmaktadır. Bitki örtüsü oranının hesaplanmasında denklem 4 kullanılmıştır:
PV=\(\left(\frac{(NDVI-NDVImin)}{(NDVImax-NDVImin)}\right)2\) (4)
where PV: Vegetation rate, NDVI: NDVI analysis, NDVImin, max: Minimum and maximum value in NDVI analysis.
After calculating the vegetation ratio, the LSE analysis was completed using the following algorithm.
where ε = Land surface emission, PV: Vegetation rate.
In the last step of the LST analysis, the analysis was completed using the following Eq. (6).
LST= \(\frac{T}{\left(1+\frac{\lambda *T1 }{c2}\right)*In\left(\right)}\) (6)
where T: Temperature degree (in Celsius), λ: Spectral Radiance wavelength (Landsat 8 band10, Landsat 7 band 6), c2: h*c/s = 1.43888*10 − 2 mK = 14,388 mK, h: Planck’s Constant:6.626*10 − 34 J s, s = Boltzmann constant = 1.38*10− 23JK, c = velocity of light = 2.998*108 m/s = 14,388, ε: LSE.
2.5 Urban Thermal Field Variance Index (UTFVI)
Urban thermal field variance indices (UTFVI) are used in the numerical evaluation of urban environments. In particular, the index used to define the urban heat island effect is calculated using the following algorithm (Zhang et al. 2006; Liu and Zhang, 2011; Luo and Wu, 2021; Cevik Degerli and Cetin, 2023).
UTFVI=\(\frac{Ts-Tmean}{Tmean}\) (7)
Where; Ts: LST, Tmean: represents the average of LST values.
There are 6 evaluation levels of calculated index values. These levels are given in Table 2.
Table 2
UTFVI index
|
UHI phenomenon
|
< 0
|
None
|
0.000-0.005
|
Weak
|
0.005–0.010
|
Middle
|
0.010–0.015
|
Strong
|
0.015–0.020
|
Stronger
|
> 0.020
|
Strongest
|
2.6 Urban LULC Indices (NDVI, NDBI, NDISI, NDWI)
Various LULC indices are used to measure the urbanization rate and urban surface characteristics (Sun et al. 2017; Xi et al. 2019; Khan et al. 2021; Sekertekin and Zadbagher, 2021; Santhosh and Shilpa, 2023).
A detailed description of the LULC indices created from the Landsat 8 OLI/TIRS satellite image is given in Table 3. ArcGIS 10.4.1 program was used to calculate the indices.
Table 3
LULC Indices
|
Formulas
|
Bands
|
NDVI (Normalized Difference Vegetation Index)
|
\(\frac{NIR-RED}{NIR+RED}\)
|
NIR = Band5
RED = Band4
SWIR1 = Band6
TIRS = Band10
GREEN = Band3
|
NDWI (Normalized Difference Water Index)
|
\(\frac{NIR-SWIR1}{NIR+SWIR1}\)
|
NDBI (Normalized Difference Built-Up Index)
|
\(\frac{SWIR1-NIR}{SWIR1+NIR}\)
|
NDISI (Normalized Difference Impervious Index)
|
\(\frac{TIRS-(MNDWI+NIR+MIR)/3}{TIRS+(MNDWI+NIR+MIR)/3}\)
MNDWI=\(\frac{GREEN-SWIR1}{GREEN+SWIR1}\)
|
NDVI values vary between − 1 and + 1. Positive values indicate the type of LULC in which vegetation is expressed, while negative values indicate residential areas, barren lands, and water bodies (Liu et al. 2017). NDVI is an important index used to detect vegetation status (Grace et al. 2007; Wilson et al. 2016; Sekertekin and Zadbagher, 2021). NDBI is an index used to determine built-up areas with satellite image data. NDBI, which has values ranging from − 1 to + 1, expresses the spatial density of built-up areas. NDBI values higher than 0.50 not only indicate a high level of construction but also a significant increase in the level of drought (Zha et al. 2003; Sekertekin and Zadbagher, 2021; Santhosh et al. 2023). NDISI is used to detect impervious surfaces (built-up areas, roads, etc.) in LULC types that create pressure on vegetation and water bodies (Sun et al. 2017; Khan et al. 2021). NDWI, on the other hand, is a remote sensing-based indicator that expresses the water resources of vegetation and land surfaces (Khan et al. 2021; Sekertekin and Zadbagher, 2021). NDWI is an index used to determine surface water resources and land cover with high humidity. It is also vital for drought monitoring, such as NDBI analysis (Rokni et al. 2014; Sekertekin and Zadbagher, 2021).
2.6 Statistical Analysis
The relationship between each of the LULC indices and LST for the years 2013, 2018, and 2023 was evaluated by linear regression analysis. Positive/negative correlations between LST and indices were determined. Analysis results were evaluated by creating regression graphs. The data required for analysis were taken from Fishnets created in the ArcGIS program with a distance of 500 * 500 m.