Geographia Polonica Vol. 94 No. 1 (2021), The localization of urban heat island in the Katowice conurbation (Poland) using the combination of land surface temperature, Normalized Difference Vegetation Index and Normalized Difference Built-up Index

The localization of Surface Urban Heat Island (SUHI) as a potential heat risk for the urban population was evaluated. The paper aimed to propose an approach to quantify and localize (SUHI) based on Landsat series TM, ETM+, OLI satellite imageries from the period 1996-2018 and recognize the Atmospheric Urban Heat Island (AUHI) effects from long term temperature measurements. Using the theoretical relation between the Normalized Difference Built-up Index (NDBI), the Normalized Difference Vegetation Index (NDVI) and the LST (Land Surface Temperature), SUHIintensity and SUHIrisk maps were created from the combination of LST, NDVI, NDBI using threshold values to localize urban heat island in the Katowice conurbation. Negative values of SUHI intensity characterize areas where there is no vegetation, highly built-up areas, and areas with high surface temperatures. The urban grow – revealed from SUHI – and global climate change are acting together to strengthen the global AUHI effect in the region as the temperature measurements were indicated.


Introduction
It is well documented that urbanization has a significant effect on local weather and climate. As a result of urbanization, an urban heat island (UHI), an unplanned change of local climate can be formed (Landsberg, 1981;Streutker, 2002;Lu et al., 2009). Greater urbanization and densely built-up areas lead to more intensive UHI, as the lower albedo and higher degree of sealing of urban areas significantly alter the surface energy budget, causing marked temperature differences between urban and rural areas.
In this context, it is important to note the difference between "atmospheric UHI (AUHI)", the occurrence of higher air temperature in urbanized areas than in surrounding areas and "surface UHI (SUHI)", the occurrence of higher surface temperature in urbanized areas than in surrounding areas (Katsoulis & Theoharatos, 1985;Oke, 1987Oke, , 1988Wang et al., 1990;Camilloni & Barros, 1997;Du et al., 2014;Kuchcik et al., 2014;Kuang et al., 2015;Majkowska et al., 2017;Kaplan et al., 2018). In Poland, during climate change, the occurrence of the AUHI or SUHI is the most important and problematic feature of urbanized areas. In Polish climatic conditions, the formation of AUHI is related to weather types connected with atmospheric fronts and summer nights where AUHI reaches higher intensity (Kłysik & Fortuniak, 1999). The AUHI in cities like Kraków can be more intense compared to other polish cities situated on relatively flat area i.e. Wrocław or Łódź. That phenomenon caused by the location of Kraków in the Vistula River valley where the katabatic flows do not enter the city interior. In the rural areas around the city, the cold air reservoirs are formed, and additionally, the sensible heat flux is significantly limited there due to the evaporation of the river waters (Bokwa, 2011). Increased land-surface temperatures caused by the SUHI effect during rapid urbanization or extension of builtup areas have a profound impact on local weather by altering local wind patterns (city structure), promoting the development of clouds and fog, increasing the number of lightning events and influencing the rates of precipitation (Oleson et al., 2001;Hung et al., 2006;Chen et al., 2009;Imhoff et al., 2010;Liu & Zhang, 2011;U.S. EPA, 2014a,b;Yang et al., 2016).
There is a theoretical relation between LST (Land Surface Temperature), the normalized difference vegetation index (NDVI) and the normalized difference built-up index (NDBI) (Qin et al., 2001;Liu & Zhang, 2011;Alfraihat et al., 2016;Kaplan et al., 2018;Gazi & Mondal, 2018). High NDVI values (ranges from -1 to 1) apply to, e.g., parks and low NDBI (ranges from -1 to 1) values with lower surface temperatures to vegetation-covered areas. Water bodies have low NDVI < 0 as well as relatively low surface temperatures (during a day) and low NDBI. Built-up areas (and bare soils) have high NDBI, low NDVI, and high surface-temperature values (Alfraihat et al., 2016;Kaplan et al., 2018). These indices can help to show the impact of green-and build-up areas on the urban heat island. The NDBI index underpins an effective technique to map builtup areas as it uses the reflectivity of buildings which is higher in the short-wave infrared band (SWIR1) and lower in the near-infrared band (NIR) (Qin et al., 2001;Alfraihat et al., 2016). Therefore, NDVI = (NIR -R)/(NIR + R) and NDBI = (NIR -SWIR1)/(NIR + SWIR1) were calculated with NIR, R (visible red bands), and SWIR.
The study area (Katowice conurbation, Poland) is an excellent place to review SUHI effects because each city in the conurbation has its own densely built-up areas. The research aimed to develop a SUHI intensity-related index (the combination of LST, NDBI, NDVI), a SUH I risk and applying these indexes to localize the potential surface urban heat islands based on threshold values. As SUHI, AUHI is related to each other and appears together within a city, the other aim was to calculate temperature duration curves -using archive temperature measurement data -to recognize the AUHI effect in the Katowice conurbation. Usually, SUHI appears within the parts of the city with a small percent of vegetation, where densely built-up areas are accompanied by high surface temperature anomalies (called heat emitters) e.g. Abegunde and Adedeji (2015), Tarawally et al. (2018) or Fashae et al. (2020). Concerning e.g. the research of Zaeemdar and Baycan (2017) who analyzed the spatial pattern (using Landsat 8 OLI) of LST, surface albedo, NDVI, and land use/land cover in Istanbul, Turkey where artificial surfaces with low albedo and low vegetation cover have the most positive exponential relationship with LST and an increasing effect on SUHI formation.

Upper Silesian conurbation
The present study is related to surface SUHI formation in the Katowice conurbation, also known as the Upper Silesian-or Silesian conurbation (Fig. 1A, B). It was established in the 1800s as one of the largest mining and industrial regions in Europe covering 3213 km² and with a population in 2013 of 2.468 million. In Central Europe, it is the second-largest urban area after Berlin. The conurbation consists of 33 cities, including the core cities of Katowice, Gliwice, Sosnowiec, Bytom, and Tychy (Krzysztofik et al., 2011). This significant commercial-and the industrialized area is based on coal mining, steel production, the mining and processing of Pb-Zn ores, coke, and textile manufacture. Between 1945Between -1989, the industrial potential of the region increased significantly. Coal mining together with other coal-based industries, i.e., iron and energy expanded (Tkocz, 2005;Spórna et al., 2016). The heavy industrialization, dense traffic, coal-burning by domestic furnaces resulting poor air quality (especially during the heating season or during summertime when wind speed is low and air-temperature is hot); the concentration of small particulate matter exceeds both daily-and yearly admissible levels (Fig. 1A, B -the air-quality data has been calculated from hourly measurements and converted to daily data for PM10). During Source of data: http://powietrze.gios.gov.pl/pjp/archives the analyzed period (2003)(2004)(2005)(2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015)(2016)(2017), the yearly PM10 concentrations reached or exceeded 50 μ μg/m 3 on ~80-120 days and 140 μ μg/m 3 on ~5-20 days every year. Even in summer, PM10 values can exceed 50 μ μg/m 3 on some days (Fig. 1B). The bad air quality is related to coal combustion (domestic) and processing, heavy industry, biomass combustion and intense traffic (Rogula-Kozłowska et al., 2011;Zaciera et al., 2012;Klejnowski et al., 2012). Usually, an unwanted effect of UHIs is the poor air quality caused by windless and cloudless conditions, especially during summertime. The cooling effect of the wind helps to eliminate the adverse effects of heat island on the micro-climate, human thermal comfort, and keep clean the air over an urbanized area (Rajagopalan et al., 2014). Furthermore, poor air quality can affect the urban quality of life and impact on human health by, e.g., aggravating asthma and causing heat-related illnesses and fatalities (Oleson et al., 2001;Liu & Zhang, 2011;Weng, 2012;U.S. EPA, 2014b;Aslan & Koc-San, 2016).

Methodology
Long-term air-quality analysis carried out at the IETU-Katowice air-quality monitoring station (GPS coordinates of the station: N: 50.264611; E: 18.975028, address: 6. Kossutha Street, Katowice). Source of raw data: http://powietrze.gios.gov.pl/pjp/archives. To obtain precise data for built-up areas, Sentinel 1 SLC data (5 images from different paths) taken from https://scihub.copernicus. eu/ were used and processed with SNAP 5.0 software. Using the average coherence and intensity VV, the VH polarization of 5 images was analyzed where low backscatter and high coherence values correspond to bare surfaces such as agricultural areas and lakes, whereas high coherence values and high backscatter values relate to built-up areas. The created RGB image (Red -the average intensity of VH, Green -the average intensity of VV, Blue -average coherence) was analyzed, using reference samples (road, water, coal-waste dump, agriculture areas, bare soils, forests, built-up areas) and the image was classified using a maximum likelihood classification. However, these data were prepared only for Figure 2 as a representative form of urban footprint in the region.
Seven Landsat images were chosen from 1996-2017 (Tab.). The image data were collected from https://earthexplorer.usgs.gov/ with zero clouds, haze, and fog cover in the summertime or at the end of spring or the beginning of autumn when there were fullgrowth vegetation and a higher risk of UHI formation. The first step in the determination of SUHI was to obtain the LST of Landsat  (4-5TM, 7 ETM+, 8 OLI) and ASTER satellite images (using thermal infrared bands (TIRS). The calculations were conducted according to https: //www.usgs.gov/land-resources/ nli/landsat/using-usgs-landsat-level-1-dataproduct. First, the radiance was calculated and then the brightness temperature was converted from Kelvin to Celsius according to T Celsius = T Kelvin -273.15 (TS = temperature at Satellite). To obtain the LST, the thermal emissivity map first had to be generated based on the spectral absorption properties of the surface materials as surface temperatures are directly related to surface physical properties: Proportion of vegetation where: NDVImin and NDVImax max = minimum-and maximum values of NDVI image (helped to separate the surface materials).

Calculating urban heat islands and risk
The combining of LST with NDVI, NDBI may help to identify those areas where SUHIs can form with different heat emissions (termed "SUHI intensity "), e.g., where there are high surface temperatures due to lack of vegetation or builtup areas like city centers with hot roofs; these which will tend to be characterized by the highest SUHI intensities. The map was calculated according to (proposed by the Author): To select those areas where heat emitters (densely built-up areas, building that transmitting thermal energy across the surface into the atmosphere) are located, the SUHI intensity map was reworked using as reference areas, the (mixed forest) code 313 from Corine Land Cover (CLC 2018) for forested areas and, for built-up urban areas, codes 111 (continuous urban), 112 (discontinuous urban), 121 (industrial-and commercial) and 122 (roadand railroad). Such reference areas were marked separately for each city on each Landsat image used. The Corine Land Cover (CLC 2018) was obtained from http://clc. gios.gov.pl. According to Martin et al. (2015) and Du et al. (2014), choosing an urban core reference area (hot surface component) and an external thermal reference (cold surface component), i.e., forested area aided the development of thresholds above the spatial average. Later, the SUHI intensity image was extracted using the CLC 2018 codes for builtup areas. Then, threshold values were developed for each extracted image and for each city to select the SUHI according to the following formula (proposed by the Author): SUHI threshold value = ((meanBUA -meanFRA) / (meanCCR -meanFRA))·STDEV BUA A further localization and classification of SUHIs called "SUHI risk " w ere prepared based on the LST image using the reference areas (forest, city center) mentioned above. The image classification was carried out according to the following formula (proposed by the Author): SUHI risk = ((LST -meanFRA Cels. ) / (meanCCR Cels. -meanFRA Cels. )) 2 The built-up areas were then extracted from the image according to the CLC 2018 codes and the results classified as follows: • <0.2 -no risk; • 0.2-0.4 -low risk; • 0.4-0.6 -moderate risk; • 0.6-0.8 -medium risk; • 0.8-1 -high risk; • >1 -highest risk.

Calculating temperature duration intervals
The Landsat series cannot give a detailed trend of air-temperature changes (fallrise), therefore a long term (last 44 years) air-temperature data set was used to see how the AUHI influences the local air-temperatures where are different albedos. Only two meteorological stations had long term available air-temperature data in the Katowice conurbation. One called "Planetarium" located in the Silesian Park -Park Śląski (GPS coordinates of the station: N: 50.290713; E: 18.992605; address: 4. Planetarium alley, Chorzów), surrounded by trees and the second is in Muchowiec Airport (GPS coordinates of the station: N: 50.240622; E: 19.032693; address: 1 Lotnisko 1, Katowice) -surrounded by meadow and buildings is located by the Katowice -Muchowiec local Airport and by a newly constructed shopping center (Fig. 6). For the temperature duration intervals, daily maximum temperatures representing the warmest part of a day and indicating the greatest extension of SUHI and AUHI with the highest temperatures were used. Daily maximum temperature data for 1973-2017 was obtained from https://ogimet. com for Muchowiec and from meteorological diaries for 1994-2016 from Chorzów Planetarium. The locations of both sites are shown in Figure 6. The collected data were grouped to construct a frequency distribution in the following way. All observations were divided into five-year intervals. Within each interval, the temperature data were ordered from the highest to the lowest without paying attention to when it was recorded in the given 5 years. The temperature data was also split into 5°C intervals. The number of data in each 5°C interval is termed as the temperature frequency index f y (h) where y denotes the fiveyear interval and h is the height (the number of data) of each 5°C interval. The accumulate frequency is calculated by the formula: where: k is the summation index.
The frequency distribution was then calculated by the formula The frequency distribution tables are presented as temperature duration curves (Fig. 7A, C).
The frequency index f y (h) shows how often in the analyzed period the daily maximum temperatures reached or exceeded the lowest level in a particular category. This cumulative frequency index can equally well be applied to characterize air temperature changes during, e.g., every 5 years.
The frequency index aided comparison of the maximum daily temperatures between 1994-2016 from Muchowiec and Chorzów Planetarium. To construct the exceedance diagram (Fig. 7E), 20°C was chosen as a fixed point and the range h y max for every five-year interval was checked to ascertain where it fell. The corresponding distribution data over those ranges where the fixed 20°C point fell was then calculated according to the formula: Results and discussion Figure 2 should be considered as an overview where Landsat image representing the morning, the ASTER image the evening state of surface temperatures (see Table -center times). The location of SUHIs in the conurbation at different parts of the day is shown in Figure 2A for daytime-and Figure 2B for night-time heat islands. Commonly, densely built-up areas, industrial areas, and bare coal-waste dumps are hot areas with temperatures of 10-12°C. Agricultural areas are 6-7°C warmer than forest and water bodies. According to Kim (1992), inner-city temperatures can be 10°C higher than those of nearby woodlands. Similar trends were reported for Polish cities like Kraków by Lewinska et al. (1982) or Bokwa (2011), from Wroclaw by Szymanowski (2004) or from Łódź by Fortuniak (2003). Water bodies are visible as heat emitters in the night-time satellite image, even warmer (4°C) then the densely built-up areas. The forested areas are also warmer (5-6°C) when compared to, e.g., agricultural areas ( Fig. 2B -northern part) or river valleys. In case of Łódź AUHI developed in the middle of the night with the greatest occurrence in the summertime (June-August) with 3-4°C, and sometimes even 6°C warmer than the surrounding areas (Kłysik & Fortuniak, 1999;Wilk, 2015). Similarly to Łódź in Białystok, the development of the AUHI starts in the afternoon and, reaching maximum in the late-evening hours (Czubaszek & Wysocka--Czubaszek, 2016).
The calculated SUHI risk , SU HI intensity, and population density for various cities of the conurbation are shown in Figure 3. Data for population density (resolution 100 meters) taken from https://www.eea.europa.eu/ data-and-maps/data/population-density-disaggregated-with-corine-land-cover-2000-2. The location of used reference areas in the formulas (5), (6) namely: meanBUA -built-up areas and meanFRA -forested areas) see on Fig. 3. Threshold values of SUHI intensity were for Katowice: 0.35; Bytom: 0.37; Gliwice: 0.35; Sosnowiec: 0.34 according to formula (5), the pixel values were considered under the threshold values. Negative values of SUHI intensity indicative of a potential SUHI location can be observed in areas with no vegetation, highly built-up areas and high LST. Furthermore, the combined image with a resolution of 30 meters can provide more detailed information concerning, e.g., roads or the shapes of hot roofs as hot surfaces, and SUHI intensity , vary within built-up areas (Fig. 3A,B). Generally, areas with the highest risk of SUHI formation occur in densely built-up areas, i.e., city centers with higher population see by Katowice, Bytom and Gliwice, industry-or logistic parks and shopping centers. The formation of stronger and size of AUHI or SUHI is depended not only on the total area of the city but also on the degree of compactness of building development in city center i.e. a high percentage of artificial surfaces, very narrow streets in a few square kilometers as it showed by Kłysik & Fortuniak (1999) in Łódź or Warsaw by Błażejczyk et al. (2016). That is well agreed with the formation of SUHIs

The localization of urban heat island in the Katowice conurbation (Poland)…
Geographia Polonica 2021, 94, 1, pp. 111-129 Figure 3A. Representative SUHI risk and SUHI intensity for Katowice and Bytom -cities from the Katowice conurbation Source of data for population density: https://www.eea.europa.eu/data-and-maps/data/population-density-disaggregated-with-corine-land-cover-2000-2 Figure 3B. Representative SUHI risk and SUHI intensity for Gliwice and Sosnowiec -cities from the Katowice conurbation Source of data for population density: https://www.eea.europa.eu/data-and-maps/data/population-density-disaggregated-with-corine-land-cover-2000-2 appearance in Bytom or Katowice where the structure of the city centers are similarly compacted as in Łódź. In the case of Sosnowiec the SUHI appearing separately and less concentrated to those areas where a dense population is (Fig. 3A,B). Therefore in the case of Katowice conurbation concerning the climate change effects, it is crucial to localize the SUHIs based on satellite images. The formation of AUHI or SUHI can promote the exposure of health risks within these densely built-up areas cconcerning the population density. Dong et al. (2014) well demonstrated a spatial-temporal pattern of higher health risk caused by UHI in urban areas, lower risk in the borderland between urban-and rural areas, and lowest risk in rural areas.
The layer stacking and averaging can eliminate several atmospheric-or surface distortions and highlighting the most problematic areas in the Katowice area. In this way, constant heat emitters such as city centers and residential areas with little or no vegetation, industrial areas, coal-waste dumps without vegetation and roads can be strengthened. The SUHI risk and SUHI intensity images were layer stacked separately and mean values calculated (Fig. 4). It is better to prepare such a layer stacking for the same summer season to avoid distortions due to land-use changes. As in the example of Du et al. (2014) who reported a long-term analysis in Nanjing (China) lasting over 20 years, most of UHIs remained in the central parts of Nanjing and their area and intensity increased remarkably over the past 20 years.
The effects of varying weather conditions on air temperatures can be easily seen in the SUHI intensity (Fig. 5, threshold values: 29.07.2013: 0.23; 30.05.2017: 0.29; 18.08.2017: 0.33; 16.09.2016: 0.34). The local air temperatures were measured between 9:00-10:00 a.m. (IETU Kossutha -same station where air quality data was measured and Muchowiec Airport meteorological station data source: https://ogimet.com) and they were compared with the LST image using mean 3x3 pixel values of ROI (Region of Interest) over the position of the station. Landsat images were acquired (scene center) between 09:30-09:40 a.m. (Tab.). The extension of the SUHI intensity significantly increased in highly built-up areas when the morning surface-and air temperatures were 31-32°C. The NDVI and NDBI can indicate places such as construction sites and bare soils which are often overheated (during daytime) and are, thus, heat emitters (Fig. 5). Also, Figure  5 well represents how the "green roof" can decrease SUHI effect and how an old structure -Spodek Arena remained as heat emitter. After the construction of the International Conference Center in Katowice has not appeared as a new SUHI close to the city center. The roof of the building complex is covered by grass with aims to create a public green place in a post-industrial area with an integrated system of land slopes and natural diversity of terrain. Such green roofing system is usually applied for energy savings by decreasing the demand for space conditioning and can help in mitigation of urban heat island effect, in the improvement of air pollution, water management, an increase of sound insulation, air humidity and its circulation, and ecological preservation (Liu & Baskaran, 2003;Berardi et al., 2014). Especially in summer, plants shade the substrate, resulting in less sunlight reaching it therefore this decreases both the absorption and the subsequent heat release. Also, green areas collect less sunlight than streets and buildings during the day (Drozd, 2019). The increasing size of such green spaces with lawns and trees can create relatively mild biothermal conditions and it can significantly reduce the SUHI effect even in a housing estate as an example showed by Błażejczyk et al. (2016) in Warsaw. The formation of new SUHI after the construction of a new shopping center is well seen on a SUHI intensity map (Fig. 6, the calculated threshold values of SUHI intensity for 23.07.1996: 0.48 and for 09.081999: 0.27). The 3 Stawy Shopping Center built in 1999 is situated close to the city center, between the A4 motorway and Dolina Trzy Stawy. The research is dealing with the localization of SUHI in Katowice conurbation. However, the application of temperature duration curves can help to introduce the AUHI effects i.e. the trend of maximal temperatures detected during summer caused by the different heat emitters (SUHI) within a built-up area. Measuring the AUHI effect is required a long term temperature data from different meteorological stations within a built-up area. The application of daily maximum-temperature duration curves at both Muchowiec (1973Muchowiec ( -2017 and Chorzów-Planetarium (1994 meteorological stations indicate increasing maximum temperatures with the passage of time at stations ( Fig. 7A-D). The Muchowiec data (Fig. 7B) reveal a huge (> 25°C) increase of temperatures between 1992-2007; this is 5% more than between 1973-1992. In the case of Chorzów--Planetarium, the temperatures with 4% more times reached and passed the 20°C border in 2014-2016 than in 1994-1998 (Fig. 7D). However, this station is located in the city park where the vegetation cover should decrease the effect of AUHI or SUHI. The duration curves can also be applied to show local changes on the exceedance diagram (Fig. 7D). Comparing the temperature exceedances > 20°C from the Muchowiec and Chorzów--Planetarium stations, a small increase (2%) at Muchowiec is evident between 1999 and 2003. Such rising maximal temperatures can be related to global climate or land-use changes. The climate change in Katowice conurbation is characterized by increases of daily average and maximum temperatures, Muchowiec 1973-1977 1978-1982 1983-1987 1988-1992 1993-1997 1998-2002 2003-2007 2008-2012 2013-2017 1973--1977 1978--1982 1983--1987 1988--1992 1993--1997 1998--2002 2003--2007 2008--2012 2013--2017 Daily max. temperatures over 25°C Daily max. temperature duration interval [%] 1994-1998 1999-2003 2004-2008 2009-2013 2014-2016 1994--1998 1999 Figure 7. Temperature exceedance duration counted for Muchowiec and Chorzów Planetarium using archived daily maximum-temperature data numbers of days with temperatures > 30°C, numbers and lengths of yearly heatwaves and numbers of days with temperature > 25°C without rain and extreme rains events resulting flash flood (Kuchcik, 2017;Fudała et al., 2018;Cenowski et al., 2019). In Poland, over 80% of nights are characterized by surplus heat in towns, amounting to 2-4°C, and sporadically to 8°C and more (Kłysik & Fortuniak, 1999). According to Kaplan et al. (2018), global climate change is expected to increase the occurrence of urban heat island effects. Thus, these urban areas may show acute economic-, social-, and environmental impacts adverse to human health. Furthermore, due to increasing numbers of inhabitants resulting from rapid urbanization, cities are more threatened by a higher risk of extreme hightemperature events such as heat waves characterized by higher intensity, higher frequency, and longer duration as a result of the UHI effect (Weng, 2003;Meehl & Tebaldi, 2004;Field et al., 2012).

Conclusions
The SUHI intensity drastically increased in highly built-up areas when the morning temperatures were 31-32°C, and the SUHI intensity is greatly influenced by seasonal-weather variations. The localization of SUHI intensity using a combination of normalized difference built-up index (NDBI), normalized difference vegetation index (NDVI), and LST were described. Negative values of SUHI intensity indicate potential SUHI (heat emitting risk areas) locations in densely built-up areas, i.e., city centers of the greatest population such as Katowice or Bytom. Therefore the SUHI intensity and SUHI risk maps confirm the application of remote sensing data in case of land-use changes and detection of SUHI phenomenon in Katowice conurbation. Such SUHI intensity and SUHI risk maps can be helpful in the development of strategies for sustainable management and preservation of urban green to moderate the SUHI environment. The results indicated well the differences between SUHI (heat emitters like densely built-up areas, construction sites) and AUHI (represented by temperature measurements). The SUHI can be well localized and reduced by increasing green space in a built-up area i.e. using green roof cover see by International Conference Center in Katowice. However, to reduce the global AUHI effect over that large built-up areas like Katowice conurbation is required more effort. The urban sprawl in the region is visible with newly formed SUHI spots like shopping centers, business parks, housing estates which will increase the AUHI effect as they emitting additional hot air to the cities' atmosphere.
Using the daily maximum temperatureduration curves well revealed the increasing temperatures in both stations which cannot be visible by the Landsat series. The tendency of increasing maximal temperatures during hot days caused probably by global climate change and urban grow which strengthening the AUHI and SUHI effect within the Katowice conurbation. A good example is showed by the meteorological data from Chorzów-Planetarium (Silesian Park -Park Śląski) where no SUHI thanks for the large vegetation cover is but the maximal temperatures are increasing in the last 20 years. Even that large vegetation cover cannot reduce the AUHI effect. That similar trend was also observed by the Muchowiec meteorological station.

Acknowledgments
This research was financed by IETU (Instytut Ekologii Terenów Uprzemysłowionych -Institute for Ecology of Industrial Areas) and it was part of a project no. 1/2017/DZM: The development of plans to adapt to climate changes in cities with over 100 thousand inhabitants. The project aimed to assess the sensitivity of the 44 largest Polish cities in the aspect of climate change and to evaluate urban planning adaptation activities.
Editors' note: Unless otherwise stated, the sources of tables and figures are the authors', on the basis of their own research.