Abstract
INTERNATIONAL JOURNAL OF URBAN SUSTAINABLE DEVELOPMENT 2021, VOL. 13, NO. 2, 248–264 https://doi.org/10.1080/19463138.2021.1872083 ARTICLE Seasonal impact of urban parks on land surface temperature (LST) in semi-arid city of Tehran a a b Farimah Sadat Jamali , Shahriar Khaledi and Mohammad Taghi Razavian a b Department of Physical Geography, Faculty of Earth Sciences, Shahid Beheshti University, Tehran, Iran; Department of Human Geography and Spatial Planning, Faculty of Earth Sciences, Shahid Beheshti University, Tehran, Iran ABSTRACT ARTICLE HISTORY Received 11 June 2020 This paper studies the impact of parks as an element of urban green infrastructure on Accepted 1 January 2021 regulating surface temperature in the semi-arid city of Tehran, Iran. Land surface temperature (LST) of parks and their immediate surroundings, derived from Landsat KEYWORDS images of winter to autumn 2017, are analysed in two urban districts with different Urban parks; land surface local climate zones and population density levels. The results show that the highest temperature; local climate levels of cooling impact are observed in spring and summer for parks larger than 1 ha zones; semi-arid; in the compact midrise setting. Although area and vegetation cover influence the sustainability impact of parks in all sizes, parks shape becomes important in parks larger than 1 ha in open built zones. To benefit from LST regulating services of park vegetation cover and facilitate achieving sustainable urban landscapes, it is recommended to opt for 1–10 ha parks, where possible, and improve vegetation cover inside parks, particu- larly in parks smaller than 1 ha. 1. Introduction variables (Ahmadi Venhari et al. 2017). These variables in regional and local scales include UGI size and Urban green infrastructure (UGI) provides cities with shape, landscape metrics, vegetation cover, location diverse ecosystem services (Martos et al. 2016; Kim within the city, built form, conditions of built-up areas and Coseo 2018; Banzhaf et al. 2019). Their climate surrounding UGI, as well as the pattern of dominant regulating services, particularly heat mitigation, has land-cover type (Cao et al. 2010; Sun et al. 2012; Lin increasingly received attention (Gunawardena et al. et al. 2015; Yan et al. 2018; Yu et al. 2018; Marando 2017; Knapp et al. 2019). This has encouraged plan- et al. 2019; Wheeler et al. 2019; Yang et al. 2020). With ning for achieving sustainability goals as well as cli- many case studies of temperate maritime (Cao et al. mate change adaptation and mitigation through 2010), humid continental (Yan et al. 2018), humid designing urban green spaces (Bowler et al. 2010; subtropical (Yu et al. 2018), mediterranean (Marando Mabon et al. 2019). UGI elements, for example, parks, et al. 2019), and oceanic climates (Yang et al. 2020), green roofs, and street trees can considerably miti- a thorough study of the role of these variables in arid gate high temperatures and cool cities in various cli- and semi-arid cities has not been carried out and thus matic regions (Jamei et al. 2016; Saaroni et al. 2018; is necessary. Marando et al. 2019; Nastran et al. 2019). The adverse The impact of UGI on regulating temperature has impacts of urban heat islands (UHI), for instance, been commonly studied through analysing air tem- increased heat vulnerability and energy consumption perature (Sodoudi et al. 2014; Eniolu et al. 2017) and (Filho et al. 2018), can be alleviated by UGI and its land surface temperature (LST) derived from remote vegetation cover. The impact of UGI elements on sensing data (Du et al. 2017; Nastran et al. 2019). The temperature is formed by internal and external CONTACT Shahriar Khaledi s-khaledi@sbu.ac.ir Department of Physical Geography, Faculty of Earth Sciences, Shahid Beheshti University, Daneshjou Boulevard, Tehran, Iran. Postcode: 1983969411 Supplemental data for this article can be accessed here. © 2021 Informa UK Limited, trading as Taylor & Francis Group INTERNATIONAL JOURNAL OF URBAN SUSTAINABLE DEVELOPMENT 249 latter has been used due to its lower cost and avail- makes daytime heat mitigation strategies necessary ability in broad spatial and temporal scales, compared (Martilli et al. 2020). Urban parks, as an element of UGI, with on-site data acquisition (Senanayake et al. 2013; can facilitate regulating temperature and accelerate Alexander et al. 2017). Although LST is important in moving towards sustainability goals (Haq 2011; the surface energy balance and affects air tempera- Emilsson and Sang 2017; Benton-short et al. 2019). ture of the lower layer of the atmosphere (Voogt and Analysing temperature regulating approaches can Oke 2003), the relationship between air temperature be improved by employing standardised methodolo- and LST varies depending on spatial scale and time of gies of classifying urban structures and functions the day (Shiflett et al. 2017). LST depends on land- (Martilli et al. 2020), for instance, the local climate cover types and the presence of pervious or imper- zone (LCZ) scheme. Mainly consisted of 10 built vious surfaces due to their different albedos, thermal types and seven land-cover types, LCZs are ‘regions properties, and latent and sensible heat fluxes (Peng of uniform surface cover, structure, material, and et al. 2012; Rogan et al. 2013; Shiflett et al. 2017; Tran human activity that span hundreds of metres to sev- et al. 2017; Choudhury et al. 2018). Vegetation land eral kilometres in horizontal scale’ (Stewart and Oke cover is associated with lower LST levels compared 2012). LCZs has characteristic temperature regimes with built-up and bare soil land covers (Weng et al. and they provide a generic classification method 2004; Alavipanah et al. 2015; Rouhi et al. 2018). based on climate for urban temperature studies Vegetation covers loss in urban areas, as well as (Stewart and Oke 2012). Although LCZ classification increased impervious surfaces, lead to higher average was initially developed for studying air temperature LST (Rogan et al. 2013). Nonetheless, as remotely (Stewart and Oke 2012; Stewart et al. 2014), it has sensed LST calculations for forests and densely tree- been shown suitable for LST and surface heat island covered areas represent the tree crowns, the surface (SUHI) analysis (Bechtel et al. 2019b; Ochola et al. temperature is possibly lower than calculated results 2020; Zhou et al. 2020) and planning for heat mitiga- (Geletič et al. 2019). tion (Das and Das 2020). Depending on the climate and the vegetation In this paper, we study the impact of urban parks in cover characteristics, the shading, reduction in surface the semi-arid city of Tehran, Iran, on regulating sur- temperature, lower heat absorption and retention, face temperature in four seasons through a year in and evapotranspiration by plants cool the air, redu- two urban districts. This is in response to the knowl- cing the risk of heat-related mortality (Shashua-Bar edge gap in quantifying the role of parks and green et al. 2009; Knight et al. 2016). Vegetation, especially spaces in LST regulation in different LCZs of the city as trees, cools the surface of hot and arid cities through well as the need for UGI planning that efficiently providing shade, evapotranspiration, intercepting addresses mitigating high daytime temperatures. solar radiation, and forming park cool island effect Like many rapidly developing cities, Tehran has (Brown et al. 2015; Shiflett et al. 2017). The cooling experienced significant built-up areas expansion and impact becomes more effective during mid-day vegetation cover decline that was accompanied by (Shashua-Bar et al. 2009; Shiflett et al. 2017) and higher LST levels in recent decades (Tayyebi et al. noticeable at the neighbourhood scale (Dialesandro 2018). This paper aims to evaluate the role of parks et al. 2019). Water-conserving ground vegetation based on their characteristics and their surroundings cover also provides cooling benefits at pedestrian while comparing the results of two urban districts (6 scale (Snir et al. 2016). Although there have been and 22) with different characteristics of local climate studies on the impact of land use and land cover on zone (LCZ) and population densities. temperature (Azhdari et al. 2018; Tayyebi et al. 2018), more studies are needed about the cooling impact of UGI specifically in developing countries and semi-arid 2. Data and methods regions (Bartesaghi Koc et al. 2018). 2.1. Study area In arid and semi-arid cities, heat islands expand at night while cool islands can be observed during day- Tehran, the capital city of Iran, is located in the south- time (Hedquist and Brazel 2014; Dialesandro et al. ern slopes of Alborz mountain ranges from 35°34ˊ 2019; Moghbel and Shamsipour 2019). However, N to 35°49ˊ N and 51°05ˊ E to 51°36ˊ E (Figure 1). these cities are faced with diurnal thermal stress that With an area of 639 km , it had an urban population 250 F. S. JAMALI ET AL. Figure 1. Boundaries of districts 6 and 22 of Tehran, with the location of synoptic stations. of more than 8.7 million in 2016 (Statistical Center of recent decades. It features a newly constructed lake Iran 2020). adjacent to an established, but degrading manmade The city has a cold semi-arid climate. Based on forest park. Both districts have faced vegetation Mehrabad synoptic station data from 1960 to 2017, cover decline over the last three decades. the annual precipitation is 212.3 mm and the mean annual temperature is 17.6°C, varying from 3.5°C in 2.2. Data January (the coldest month) to 30.9°C in July (warmest month). Landsat 8 Operational Land Imager (OLI) and Thermal Considering their differences and similarities, dis- Infrared Sensor (TIRS) daytime images (path 164 and tricts 6 and 22 of Tehran may provide suitable case row 35) were acquired from USGS Earth Explorer for studies for LST analysis with implications for sustain- the following dates in 2017:30 January 2022 May, able urban design. District 6 (D6) has been estab- 25 July and 29 October. Each day was selected as lished for over six decades in the heart of the city. a representative day in each season of 2017. Table 1 A population of more than 251,000 persons is spread shows the properties of images on the acquisition over an area of 21.4 km in D6. District 22 (D22) is at dates. Table 2 shows the corresponding climate the westernmost part of the city. It has an area of records from Geophysics (D6, 35°45ˊ N, 51°23ˊ E) and 58.5 Km with a population of more than 176,000 per- Chitgar (D22, 35°44ˊ N, 51°10ˊE) synoptic stations. The sons. This district has experienced rapid develop- images underwent preprocessing that includes atmo- ment, land-use change, and population growth in spheric correction based on DOS1 method (Chavez Table 1. Landsat images used for LST studies. Time Date Landsat Scene ID Time (GMT) (Local) Sun Azimuth Angle Sun Elevation Angle Earth – Sun Distance 30-January LC81640352017030LGN00 07:08:09 10:38:09 152.639 31.682 0.985 22-May LC81640352017142LGN00 07:07:37 11:37:37 126.036 66.432 1.012 25-July LC81640352017206LGN00 07:08:01 11:38:01 123.692 64.263 1.0157 29-October LC81640352017302LGN00 07:08:24 10:38:24 159.142 37.9415 0.993 INTERNATIONAL JOURNAL OF URBAN SUSTAINABLE DEVELOPMENT 251 Table 2. Climate records of Chitgar (Chi.) and Geophysics (Geo.) stations on the days of Landsat images retrieval. Date 30 January 2017 22 May 2017 25 July 2017 29 October 2017 Station Chi. Geo. Chi. Geo. Chi. Geo. Chi. Geo. Mean Air Temperature (°C) 1.5 1.2 23.5 22.0 31.9 25.1 22.8 21.1 Max. Air Temperature (°C) 6.1 5.7 28.8 27.8 37.0 30.5 28.2 27.3 Min. Air Temperature (°C) −2.8 −3.3 16.6 14.6 24.6 18.9 18.4 16.6 Rainfall (mm) 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Figure 2. Urban parks and street networks in D6 (right) and D22 (left). −2 −1 −1 1996) and visible bands transformation to reflectance. radiance (W m sr μm ) and BT is brightness tem- Figure 2 shows the urban parks in D6 (38 parks) and perature (°C). The proportion of vegetation (P ) is D22 (40 parks). calculated from Equation (2) and then the land surface emissivity (ɛ) is estimated (Equation 3): 2.3. Land-cover classification and LST � � NDVI NDVI min calculation Pv ¼ (2) NDVI NDVI max min Through supervised classification and using SCP plu- gin in QGIS, an open-source geographic information system program, the urban area was classified to four ε ¼ 0:004Pvþ 0:986 (3) land-cover macro-classes, including water, vegetation, built-up, and bare soil (Congedo 2016). where NDVI and NDVI denote the minimum and min max LST retrieval was based on calculating the bright- maximum NDVI values of each image, respectively. ness temperature of thermal bands and estimating Subsequently, the land surface temperature (°C) is land surface emissivity from normalised difference calculated from the following equation: vegetation index (NDVI) (Sobrino et al. 2004; Avdan and Jovanovska 2016). Accordingly, the brightness BT hc LST ¼ ;ρ ¼ (4) temperature is calculated from: 1þðλBT=ρÞ lnðεÞ k � � BT ¼ 273:15 (1) where λ is the mean wavelength of the thermal band K1 ln 1þ (μm), ρ equals 0.01438 mK, h is the Planck’s constant, −2 −1 −1 c is the speed of light, and k is the Boltzmann constant. where K is 774.89 W m sr μm and K is 1 2 1321.08 K for Landsat 8 – band 10, L is spectral λ 252 F. S. JAMALI ET AL. 2.4. Urban characteristics positive values indicate their cooling impact. It should be noted that since the calculated LST values are The difference between D6 and D22 is highlighted by based on the urban surfaces seen by remote sensors their population density and proportion of dominant (Voogt and Oke 2003), the studied scale does not urban patch (Guo et al. 2018) as well as LCZs. LCZ map address the variations at smaller scales and below of the city is produced using WUDAPT Level 0 meth- the surfaces, for instance, at a pedestrian level below odology (Bechtel et al. 2015, 2019a) based on LCZ tree crowns. types (Stewart and Oke 2012). The workflow includes resampling Landsat data of the region of interest, here the image of 25 July 2017, from 30 m to 100 m pixels, 2.6. Urban parks characteristics digitising training areas for each LCZ in Google Earth, and implementing supervised classification of the The quantitative characteristics of parks, including the Landsat data in SAGA-GIS (Bechtel et al. 2015). The proportion of vegetation cover inside the park (%) classification results are assessed by calculating over- derived from land-cover classification, park area (ha) all accuracy and Kappa coefficient using a pixel-based and landscape shape index (LSI) were calculated. LSI is confusion matrix. calculated using Equation (6) (Cao et al. 2010): LSI ¼ pffiffiffiffiffiffi (6) 2 πA 2.5. Impact of urban parks on temperature where P is the total perimeter (m) and A is the area of The impact of urban parks in cold and warm seasons a park (m ). are studied. The impact can be defined by the differ - Since a few of the parks had waterbodies that do ence between the LST of a park and its surrounding not provide statistically significant analyses, these fea- area (Cao et al. 2010; Lin et al. 2015; Di Leo et al. 2016; tures were not considered in this study. Yu et al. 2018): To study the impact of parks, the correlation between the characteristics and the following vari- LST ¼ LST LST (5) u p ables was analysed in different seasons: mean LST of where LST and LST represent the average LST of the park (°C), warming/cooling extent (m) and impact p u −1 a park and its surrounding urban area, respectively. (°C) as well as surface temperature gradient (°C km ). This can show the park cool island (PCI) intensity of Due to the wide range of park areas, the logarithmic a park (Bartesaghi Koc et al. 2018). Accordingly, we scale was used to assess the correlation of park area calculated the mean LST of urban parks and their with other variables. surroundings within 500 m buffer zones in 10 m seg- The analysis was first accomplished for all parks in ments from the edge of parks. The mean LST of each D6 and D22. Subsequently, the parks were classified segment was plotted versus distance. The tempera- based on their area in 3 groups of 0–1 ha, 1–10 ha and ture regulating impact of UGI diminishes with dis- larger than 10 ha to clarify the role of their size. tance (Bowler et al. 2010; Yu et al. 2018). Therefore, the LST regulating distance of a park was estimated by the distance between the edge of the park and the 3. Results first turning point of LST drop, or until the LST reached 3.1. Land cover, LCZ and LST a constant level. The LST regulating impact (cooling/ warming) was calculated as the LST difference Table 3 shows the proportion of four mainland cover between the park and the chosen section (Yu et al. types (water, built-up, vegetation, and bare soil) in D6 2018). The negative values of temperature regulation and D22 in July 2017. Built-up is the dominant land impact of parks show their warming impact, while cover in both districts. The total proportion of Table 3. Areas of land-cover macro-classes in the summer 2017 and population density in D6 and D22. −2 Total Area (ha) Water (ha) % Built-up (ha) % Vegetation (ha) % Bare Soil (ha) % Pop. Density (persons km ) D6 2142.94 0.0 0.0 1868.9 87.2 237.3 11.1 36.8 1.7 11,731 D22 5855.72 130.7 2.2 3263.8 55.7 874.8 14.9 1586.4 27.1 3012 INTERNATIONAL JOURNAL OF URBAN SUSTAINABLE DEVELOPMENT 253 Figure 3. Local climate zones (LCZs) of D6 (right) and D22 (left). vegetation cover (%) in January, May, and LCZ in D6, followed by open midrise and small October 2017 were 4.2, 10.6, and 5.5 in D6, and 7.5, patches of scattered trees. Large low-rise settings, 14.9, and 8.4 in D22, respectively. Land-cover classifi - including commercial and research facilities as well cation maps derived from Landsat images on selected as large construction sites, along with open high, mid, dates of 2017 can be found in Figure S1 of the online and low-rise LCZs are typical built types in D22. supplements. Figure 3 shows the LCZ map of D6 and Compared with D6, D22 encompasses extensive nat- D22 produced based on WUDAPT L0 method with ural land cover of bare soil and scattered trees. a majority filter of 2 pixels (200 m). The overall accu- Figure 4 illustrates the LST variations throughout racy results for the land-cover maps of January, May, the year. It shows that levels of LST in the compact July, and October as well as LCZ are 98.6, 97.7, 97.0, setting of D6 are considerably lower than D22, parti- 97.3, and 83.6, respectively. The corresponding Kappa cularly in southern and western parts with large low- coefficients are 0.97, 0.96, 0.94, 0.95, and 0.79, respec- rise and nonvegetated land cover. The mean LST in tively. Table 4 shows the profile of D6 and D22 in January, May, July, and October in D6 is 3.4°C, 33.4°C, terms of LCZ types. Compact midrise is the dominant 37.0°C, and 23.0°C, respectively. Similarly, they are 5.6° C, 36.7°C, 41.1°C , and 26.5°C in D22, respectively. Table 5 shows the relationship between parks Table 4. Summary of local climate zone proportions in D6 and D22 internal and external variables. A general correlation (%). is observed between park area and vegetation cover LCZ D6 D22 with their impact and corresponding distance. 1 Compact high-rise 0.0 0.0 Stronger correlations and statistically significant linear 2 Compact midrise 71.8 0.4 3 Compact low-rise 0.0 0.1 relationships between parks internal variables and 4 Open high-rise 1.3 7.5 cooling distance and impact are identified in warmer 5 Open midrise 19.7 9.3 seasons. This pattern is also observed for LST of the 6 Open low-rise 0.2 7.0 7 Lightweight low-rise 0.0 0.0 parks which is negatively correlated with vegetation 8 Large low-rise 2.2 21.0 cover in warm seasons. Shape index does not corre- 9 Sparsely built 1.3 2.5 late with park impact in D6 while it becomes impor- 10 Heavy industry 0.0 0.0 A Dense trees 0.1 0.0 tant throughout the year in D22. B Scattered trees 2.6 16.6 C Bush, scrub 0.0 3.5 D Low plants 0.0 1.2 3.2. Impacts on LST based on park size E Bare rock or paved 0.3 11.0 F Bare soil or sand 0.4 17.0 Figure 5 shows the profile of park variables based on G Water 0.0 2.8 their size in D6 and D22. The areas of parks in D6 254 F. S. JAMALI ET AL. Figure 4. LST maps of selected days in D6 (right) and D22 (left). INTERNATIONAL JOURNAL OF URBAN SUSTAINABLE DEVELOPMENT 255 Table 5. Pearson correlation coefficients and linear regression F-test R-squared (* shows p < 0.05). Log Area (ha) LSI Vegetation Proportion (%) 2 2 2 District Cor. R Cor. R Cor. R Winter 6 LST Park 0.267 0.071 0.055 0.003 −0.091 0.008 Distance 0.799 0.639* −0.057 0.003 0.768 0.590* Impact −0.152 0.023 −0.077 0.006 −0.144 0.021 LST Grad. 0.129 0.017 −0.210 0.044 0.195 0.038 22 LST Park 0.313 0.098* 0.243 0.059 0.192 0.037 Distance 0.876 0.768* 0.491 0.242* 0.506 0.256* Impact 0.404 0.163* 0.214 0.046 0.336 0.113* LST Grad. 0.236 0.056 0.204 0.042 0.273 0.074 Spring 6 LST Park −0.489 0.239* 0.103 0.011 −0.590 0.348* Distance 0.790 0.623* 0.153 0.023 0.732 0.536* Impact 0.809 0.655* 0.060 0.004 0.613 0.376* LST Grad. 0.494 0.244* −0.001 0.000 0.328 0.108* 22 LST Park 0.217 0.047 0.086 0.007 −0.314 0.099* Distance 0.823 0.677* 0.552 0.304* 0.367 0.135* Impact 0.578 0.334* 0.465 0.216* 0.579 0.336* LST Grad. −0.023 0.001 −0.121 0.015 0.516 0.266* Summer 6 LST Park −0.536 0.287* 0.004 0.000 −0.630 0.396* Distance 0.745 0.555* 0.048 0.002 0.618 0.382* Impact 0.811 0.658* 0.034 0.001 0.620 0.384* LST Grad. 0.456 0.208* −0.020 0.000 0.315 0.099 22 LST Park 0.225 0.050 0.033 0.001 −0.381 0.145* Distance 0.818 0.669* 0.552 0.305* 0.207 0.043 Impact 0.528 0.279* 0.380 0.145* 0.561 0.315* LST Grad. −0.001 0.000 −0.065 0.004 0.580 0.337* Autumn 6 LST Park −0.160 0.026 0.044 0.002 −0.264 0.070 Distance 0.827 0.684* 0.041 0.002 0.733 0.538* Impact 0.599 0.359* 0.049 0.002 0.414 0.172* LST Grad. −0.103 0.011 0.136 0.018 −0.116 0.014 22 LST Park 0.343 0.118* 0.198 0.039 −0.064 0.004 Distance 0.849 0.721* 0.596 0.355* 0.291 0.085 Impact 0.213 0.045 0.078 0.006 0.375 0.140* LST Grad. −0.094 0.009 −0.203 0.041 0.268 0.072 range from 0.04 ha, the smallest, to 27.04 ha, the larger parks show more extensive impact. The regulat- largest, and in D22 from 0.14 ha to 979.58 ha. ing distance is relatively higher in D22. Mean tempera- −1 Compared with D22, proportions of vegetation cover ture gradients in the spring are 10.6°C km in D6 and −1 in D6 parks larger than 1 ha are higher throughout 10.7°C km in D22. During the summer, temperature the year, especially in spring and summer. The linear regulation in shorter distances produces high-tem- −1 and peripheral layout of several parks in D22 gener- perature gradients that on average are 10.7°C km −1 ates larger LSI values compared with D6. in D6 and 10.°C km in D22. In some cases, particu- Table 6 shows the mean, maximum, and minimum larly in parks smaller than 1 ha, the temperature gra- −1 LST regulating the impact of urban parks in D6 and dients reach beyond 20°C km . D22 based on their size group. Table 7 shows the Figures 6–8 show the relationship between vege- corresponding distances. A general cooling impact is tation cover inside the parks and their impact on LST. observed in the spring, summer, and autumn. As the The Pearson correlation coefficients between parks cooling impact increases in spring, the maximum characteristics and impact variables based on their cooling of parks larger than 10 ha in D6 reaches 4.6° size groups can be found in Table S1 of the online C, 1.9°C more than D22 (2.7°C), whilst the maximum supplements. Considering the parks based on their cooling distance in D22 extends to 270 m, compared size shows that, similar to the results of Table 5, LST with 210 m in D6. The temperature regulating dis- of the parks are negatively correlated with park areas tance remains below 270 m throughout the year and and vegetation cover proportions. LST regulation 256 F. S. JAMALI ET AL. Figure 5. Boxplot of parks internal variables based on size (0-1 ha, left; 1-10 ha, centre; > 10 ha, right) in D6 and D22. INTERNATIONAL JOURNAL OF URBAN SUSTAINABLE DEVELOPMENT 257 Table 6. Overview of LST regulating the impact of parks (°C) based on park area. District 6 22 Area (ha) 0–1 1–10 >10 Total 0–1 1–10 >10 Total Season No. 27 8 3 38 21 16 3 40 Winter Mean −0.3 −0.6 −0.1 −0.3 −0.1 −0.4 0.7 −0.2 Max. 0.0 −0.2 0.9 0.9 0.3 0.3 1.3 1.3 Min. −1.2 −1.0 −0.9 −1.2 −0.5 −1.1 −0.1 −1.1 Spring Mean 0.4 1.2 3.5 0.8 0.5 0.8 1.7 0.7 Max. 1.5 2.2 4.6 4.6 1.8 2.7 2.7 2.7 Min. −0.1 0.6 2.8 −0.1 −0.2 −0.4 0.5 −0.4 Summer Mean 0.4 1.1 3.2 0.8 0.5 1.1 1.5 0.8 Max. 1.6 1.8 4.5 4.5 1.7 3.0 2.5 3.0 Min. 0.0 0.2 2.2 0.0 −0.1 −0.1 0.5 −0.1 Autumn Mean 0.2 0.3 1.0 0.3 0.3 0.4 0.6 0.4 Max. 0.6 0.5 1.7 1.7 3.0 1.2 0.9 3.0 Min. 0.0 0.0 0.6 0.0 0.0 −0.6 0.3 −0.6 Table 7. Overview of LST regulating distance of parks (m) based on park area. District 6 22 Area (ha) 0–1 1–10 >10 Total 0–1 1–10 >10 Total Season No. 27 8 3 38 21 16 3 40 Winter Mean 34 91 153 56 32 53 180 52 Max. 60 120 180 180 70 80 210 210 Min. 10 40 100 10 10 30 130 10 Spring Mean 50 88 157 67 40 64 187 61 Max. 110 140 210 210 70 160 270 270 Min. 20 50 130 20 10 10 140 10 Summer Mean 46 91 160 64 50 78 187 71 Max. 110 140 220 220 80 160 270 270 Min. 10 50 130 10 10 30 140 10 Autumn Mean 29 65 133 45 36 57 143 52 Max. 50 100 170 170 70 110 210 210 Min. 10 20 100 10 10 20 100 10 impact is correlated with vegetation cover in almost the introduction of vegetation cover in arid cities can all park size groups and throughout the year. The make them cooler than their surroundings covered statistically significant (p < 0.05) coefficients of deter- with bare soil (Fathi et al. 2019). mination in May and July, in Figure 7, support the role The results of this study show that park character- of the proportion of vegetation cover in LST cooling istics influence LST throughout the year. Generally, impact, particularly in 1–10 ha parks. parks are cooler than their surrounding in warm sea- sons. As shown in Table 5, the results indicate the correlation between park areas and the proportion 4. Discussion of vegetation covers inside the parks with cooling 4.1. Parks impacts and implications for impact becomes more prominent in the spring and sustainable urban landscapes summer, with a statistically significant linear relation- ship in both districts. This is similar to the results of the The results of LST calculation through seasons of 2017 study on the relationship between NDVI of parks and show higher LST in D22 than D6 (Figure 4), although LST in a semi-arid region of California (Dronova et al. D22 has a relatively greater proportion of waterbodies. 2018). The pronounced seasonal correlation can be This could be explained by the higher proportion of attributed to the general-increased level of vegetation bare soil in D22 and large low-rise developments with cover, compared with the cold season, and the asso- minimal vegetation land cover. Yet, open high and ciated cooling due to the higher albedo of tree mid-rise LCZs, that are consisted of low plants and crowns, shading, and evapotranspiration (Geletič scattered trees (Stewart and Oke 2012), in eastern et al. 2019; Jamshidi et al. 2019). parts of D22 are accompanied by lower LST levels, as 258 F. S. JAMALI ET AL. Figure 6. Relationship between LST regulation impact and vegetation cover in D6 (top) and D22 (bottom) for 0-1 ha parks on selected dates of 2017. Figure 7. Relationship between LST regulation impact and vegetation cover in D6 (top) and D22 (bottom) for 1-10 ha parks on selected dates of 2017. The calculated LST of parks in the winter is influ - properties including albedo and heat storage. For enced by the relatively lower proportion of vegetation several parks in the winter, the mean LST inside the cover, the dominance of leafless deciduous plants, park is lower than their surroundings, while the and bare soil and their corresponding thermal reverse is seen for others. Although this can be INTERNATIONAL JOURNAL OF URBAN SUSTAINABLE DEVELOPMENT 259 Figure 8. Relationship between LST regulation impact and vegetation cover in D6 (top) and D22 (bottom) for parks larger than 10 ha on selected dates of 2017. connected to their area and type of dominant land semi-arid cities. The results shown in Table 7 indicate cover, the relationship is not supported by significant that the LST regulating impact of parks and the vegeta- correlation coefficients. This has produced both cool- tion cover in the semi-arid city of Tehran occurs on ing and warming impacts results in the winter. For a local scale, particularly for parks smaller than 1 ha. example, the largest park in the study area, that fea- This is observed both in densely populated compact tures a 111.69 ha manmade lake and a mix of conifer- built type with small parks and lower density neigh- ous and broadleaf tree cover in D22, depicts a cooling bourhoods with open built LCZ types. The LST turning impact throughout the year. points that define the regulating impact in both dis- Among the internal variables, area and vegetation tricts have occurred in distances less than 270 m. This is cover play important roles in LST regulation in both similar to the average range of daytime cooling extents districts. When considering the parks in general, LSI of case studies from small urban parks in Addis Ababa only appears important in D22 and it is positively (Feyisa et al. 2014), Beijing (Lin et al. 2015), Melbourne correlated with a cooling impact. Yet, classifying the (Al-Gretawee et al. 2016) as well as the results of the parks based on their areas reveals that LSI becomes an study on 10 dryland cities (Dialesandro et al. 2019). influential factor in the impact of parks larger than 1 ha; However, the maximum cooling distances in these similar to the results of the study on green spaces in studies reach beyond 400 m (Feyisa et al. 2014) and Leipzig (Jaganmohan et al. 2016). LSI becomes notable 800 m (Lin et al. 2015; Al-Gretawee et al. 2016). It in the process of designing urban parks in open LCZ explains that the role of vegetation cover in cooling settings and relatively less densely populated districts arid and semi-arid cities can be prominent at the neigh- with the possibility of creating parks larger than 1 ha. bourhood scale, rather than city-wide scale Based on the results shown in Table 6, the LST (Dialesandro et al. 2019). regulation impact reaches its highest during the spring Although the cooling extent of urban forests in (May) and summer (July), creating considerable LST some dryland cities can reach more than 1 km gradients between parks and their surroundings that (Dialesandro et al. 2019), the results of analysing −1 can go beyond 10°C km in both districts. It empha- parks in D22 of Tehran show that large urban parks sises the role of parks and vegetation cover in cooling with low proportions of vegetation cover do not 260 F. S. JAMALI ET AL. produce extensive LST cooling impacts. When consid- it is recommended that the existing tree cover to be ering all the parks in both districts, the mean LST maintained. Due to high water consumption of orna- cooling impact reaches 0.8°C with a mean regulating mental grasses and low height vegetation cover with distance of 68 m in the summer. These values are insignificant shadow, more attention should be paid considerably lower than the results of studies in to improving tree cover and using drought-tolerant other climatic regions (Yu et al. 2018; Guo et al. species. Grouping the vegetation, particularly tree 2019; Marando et al. 2019; Yang et al. 2020). cover, in clusters may help in effective LST regulation Nevertheless, the obtained results from remote sen- (Shiflett et al. 2017). sing at local scale do not overlook the significant Tehran has a high population density and built-up impact of vegetation cover, particularly trees, at street areas that play a part in warming the city. It also faces and neighbourhood scale microclimates. the challenges produced by the impacts of climate The cooling impact of parks larger than 1 ha in D6 change (Alizadeh-Choobari et al. 2016). These issues is higher than D22. This can be due to the larger necessitate acknowledging and optimising the ecosys- proportion of vegetation cover in parks larger than tem services provided by the UGI elements of the city 1 ha in D6, illustrating the role of densely vegetated through the process of planning for the sustainable areas in LST regulation. The dominant LCZ type in D6, urban landscape and climate change adaptation. The compact mid-rise (LCZ 2), can also play a role in results of this study show that urban parks provide the creating noticeable PCI effect between a park and its city with daytime LST cooling impact during warm adjacent urban area. It should be noted that there are seasons. This is particularly vital for the heat vulnerabil- a few parks larger than 10 ha in D6 and D22. This ity experienced by residents in arid and semi-arid cities limited the statistical significance of the correspond- (Jenerette et al. 2016; Martilli et al. 2020). ing results for linear regression. Since parks larger than 10 ha, compared with 4.2. The research limitations and 1–10 ha parks, do not necessarily lead to substantially recommendations larger and more extensive cooling impacts, the 1–10 ha parks are valuable in regulating LST in Although remotely sensed images provide valuable Tehran, in both districts with different population data to estimate surface parameters (Weng 2009) in density, built-up and LCZ types. However, the short- various spatial and temporal scales, they introduce age of space may limit new park development in limitations to the study. The limits arise from the neighbourhoods with compact LCZ types. This issue partial view of urban surfaces with areal fractions of could be addressed by enhancing vegetation cover ground cover, rooftops and vegetation canopies inside the parks smaller than 1 ha, due to the correla- (Voogt and Oke 2003; Bechtel et al. 2019b), the impact tion between vegetation cover and temperature reg- of latitude and data retrieval time on thermal aniso- ulation impact. It is suggested that the number of tropy due to nonhomogeneous heating of urban sur- small parks should be increased in high-density faces (Krayenhoff and Voogt 2016), and the inability of areas. This supports climate regulation as well as cul- remote observations in capturing shortwave radiation tural and social services of parks, as the city lacks the converting to evaporation latent heat (Yu et al. 2018). appropriate distribution of parks and green spaces Moreover, this study does not consider air tem- (Bahrini et al. 2017). Vegetation cover, particularly perature. Although LST modulates air temperature of trees with shadow, can be effective in alleviating the lower atmosphere (Voogt and Oke 2003), and its heat in warm and arid regions (Hedquist and Brazel analysis can describe the daytime cooling impact of 2014; Wang et al. 2015; Dialesandro et al. 2019; vegetation cover on surface temperature and facili- Marando et al. 2019; Wheeler et al. 2019). The heat- tate studying PCI (Cao et al. 2010; Du et al. 2017; Fan regulating the impact of trees can be increased by et al. 2019), the results of the study cannot be directly using proper species in optimal locations within the interpreted as thermal comfort, that can be described urban environment, for instance, trees with high-den- by a combination of parameters such as air tempera- sity foliage that provides shade in open settings or ture, relative humidity, and vapour pressure places with low sky view factors (Gillner et al. 2015; (Mahmoud and Gan 2018). Morakinyo et al. 2020). Since available water resources Since the findings of the paper are based on day- are limited in warm seasons, particularly for irrigation, time Landsat images, the night-time impact of parks INTERNATIONAL JOURNAL OF URBAN SUSTAINABLE DEVELOPMENT 261 is not addressed. The results of previous studies with planning for small but abundantly vegetated about the night-time impact of urban vegetation parks may reduce adverse impacts of the daytime heat. cover on air temperature or LST vary depending on the vegetation cover types, climate, urban form, Disclosure statement scale, etc. For instance, grass and tree covers can provide different cooling impacts at micro to regio- In accordance with Taylor & Francis policy and their ethical obligation as researchers, the authors declare that they have nal scales on air and surface temperature (Shiflett no conflict of interest. et al. 2017). Tree canopies can hinder long-wave radiative cooling at night that can lead to increased air temperature in low canopies (Wheeler et al. 2019), Notes on contributors decreased vegetation covers cooling impact com- Farimah Sadat Jamali is a Water Resources Engineering and pared with daytime cooling (Zhang et al. 2017) or Landscape Architecture graduate and she currently carries out negligible cooling of vegetation covers areas in dry- her PhD thesis in the field of urban climatology at Shahid land cities (Dialesandro et al. 2019). Accordingly, Beheshti University, Tehran, Iran. She studies the contribution evaluating the night-time behaviour of vegetation of urban green and blue infrastructure and nature-based solu- tions to sustainable development and climate change mitiga- cover on LST can provide a comprehensive overview tion and adaptation. of park impact. Furthermore, this paper employed Shahriar Khaledi holds a PhD in Climatology and Environmental the proportion of vegetation cover as a variable. Planning and he is a professor of Climatology at Shahid Beheshti Some studies included the analysis of NDVI (Azhdari University, Tehran, Iran. His research interests include urban et al. 2018; Dronova et al. 2018). In future studies, it is climatology, urban environmental planning, and climate suggested considering the LST variations in relation- change. He has supervised many masters’ and doctoral theses ship with detailed characteristics of vegetation on physical geography and climatology. He has authored sev- eral books and his papers have been published in scientific cover. Studying the impact of the spatial distribution journals and conference proceedings. of parks within the districts is also recommended. Mohammad Taghi Razavian holds a PhD in Human Geography and he is a professor of Urban Planning at Shahid Beheshti 5. Conclusion University, Tehran, Iran. His field of research encompasses sus- tainable urban development, urban planning, and urban envir- This paper studied the influential characteristics of onment. He has supervised several masters’ and doctoral theses urban parks on regulating land surface temperature on human geography and urban planning. He is the author of (LST) in two different districts of Tehran (districts 6 several books as well as he has published papers in scientific journals. and 22) in terms of population density, proportion of built-up areas and local climate zone types. The results demonstrate the significant role of vegetation cover on References LST regulation inside and outside the parks in both Ahmadi Venhari A, Tenpierik M, Mahdizadeh Hakak A. 2017. districts, particularly in warm seasons. Similarly, LST Heat mitigation by greening the cities, a review study. regulation impacts and distances are correlated with Environ Earth Ecol. 1(1):5–32. doi:10.24051/eee/67281. park areas. Shape index is an influential factor in parks Alavipanah S, Wegmann M, Qureshi S, Weng Q, Koellner T. 2015. with areas larger than 1 ha in both districts. Compared The role of vegetation in mitigating urban land surface tem- with studies in other climate regions, the impact of the peratures: a case study of Munich, Germany during the warm season. Sustainability. 7(4):4689–4706. doi:10.3390/ parks in the semi-arid city on LST is relatively local, su7044689. forming high surface temperature gradients and park Alexander PJ, O’Dwyer B, Brennan M, Mills G, Lynch P 2017. Land cool island intensity between the parks and their sur- surface temperature climatology over urban surfaces: roundings. The results suggest that parks with higher a blended approach. Proceedings of the Joint Urban vegetation cover and areas between 1 and 10 ha can Remote Sensing Event (JURSE); Mar 6–8; Dubai, United Arab be more efficient, than smaller or larger parks, in reg- Emirates: IEEE; p. 1–4. Al-Gretawee H, Rayburg S, Neave M. 2016. The cooling effect of ulating land surface temperature in districts with domi- a medium sized park on an urban environment. nant compact mid-rise local climate zone. However, in Int J GEOMATE. 11(26):2541–2546. highly populated urban districts with limited space for Alizadeh-Choobari O, Ghafarian P, Adibi P. 2016. Inter-annual urban green infrastructure development, maintaining variations and trends of the urban warming in Tehran. Atmos and improving the existing vegetation cover along Res. 170:176–185. doi:10.1016/j.atmosres.2015.12.001. 262 F. S. JAMALI ET AL. Avdan U, Jovanovska G. 2016. Algorithm for automated map- Congedo L 2016. Semi-automatic classification plugin ping of land surface temperature using Landsat 8 satellite documentation. Release 5.3.2.1. [accessed 2019 Jan 9]. data. J Sensors. 2016:1–8. doi:10.1155/2016/1480307. http://dx.doi.org/10.13140/RG.2.2.29474.02242/1 Azhdari A, Soltani A, Alidadi M. 2018. Urban morphology and Das M, Das A. 2020. Assessing the relationship between local landscape structure effect on land surface temperature: evi- climatic zones (LCZs) and land surface temperature (LST)– dence from Shiraz, a semi-arid city. Sustain Cities Soc. a case study of Sriniketan-Santiniketan Planning Area 41:853–864. doi:10.1016/j.scs.2018.06.034. (SSPA), West Bengal, India. Urban Clim. 32:100591. Bahrini F, Bell S, Mokhtarzadeh S. 2017. The relationship doi:10.1016/j.uclim.2020.100591. between the distribution and use patterns of parks and Di Leo N, Escobedo FJ, Dubbeling M. 2016. The role of urban their spatial accessibility at the city level: a case study from green infrastructure in mitigating land surface temperature Tehran, Iran. Urban For Urban Green. 27:332–342. in Bobo-Dioulasso, Burkina Faso. Environ Dev Sustain. 18 doi:10.1016/j.ufug.2017.05.018. (2):373–392. doi:10.1007/s10668-015-9653-y. Banzhaf E, De la Barrera F, Reyes-paecke S. 2019. Urban green Dialesandro JM, Wheeler SM, Abunnasr Y. 2019. Urban heat infrastructure in support of ecosystem services in a highly island behaviors in dryland regions. Environ Res Commun. 1 dynamic South American city: a multi-scale assessment of (8):1–12. doi:10.1088/2515-7620/ab37d0. Santiago de Chile. In: Schröter M, Bonn A, Klotz S, Seppelt R, Dronova I, Friedman M, McRae I, Kong F, Yin H. 2018. Spatio- Baessler C, editors. Atlas of ecosystem services- drivers, risks, and temporal non-uniformity of urban park greenness and ther- societal responses. Cham: Springer Nature Switzerland; p. mal characteristics in a semi-arid region. Urban For Urban 157–165. Green. 34:44–54. doi:10.1016/j.ufug.2018.05.009. Bartesaghi Koc C, Osmond P, Peters A. 2018. Evaluating the Du H, Cai W, Xu Y, Wang Z, Wang Y, Cai Y. 2017. Quantifying the cooling effects of green infrastructure: a systematic review cool island effects of urban green spaces using remote sen- of methods, indicators and data sources. Sol Energy. sing data. Urban For Urban Green. 27:24–31. doi:10.1016/j. 166:486–508. doi:10.1016/j.solener.2018.03.008. ufug.2017.06.008. Bechtel B, Alexander PJ, Beck C, Böhner J, Brousse O, Ching J, Emilsson T, Sang ÅO. 2017. Impacts of climate change on urban Demuzere M, Fonte C, Gál T, Hidalgo J, et al. 2019a. areas and nature-based solutions for adaptation. In: Generating WUDAPT level 0 data- current status of produc- Kabisch N, Korn H, Stadler J, Bonn A, editors. Nature-based tion and evaluation. Urban Clim. 27:24–45. doi:10.1016/j. solutions to climate change adaptation in urban areas. Cham: uclim.2018.10.001. Springer Nature; p. 15–27. Bechtel B, Alexander PJ, Böhner J, Ching J, Conrad O, Feddema J, Eniolu T, Dahanayake KWDKC, Ng E, Chow CL. 2017. Mills G, See L, Stewart I. 2015. Mapping local climate zones for Temperature and cooling demand reduction by green-roof a worldwide database of the form and function of cities. types in different climates and urban densities: a ISPRS Int J Geo-Inf. 4(1):199–219. doi:10.3390/ijgi4010199. co-simulation parametric study. Energy Build. 145:226–237. Bechtel B, Demuzere M, Mills G, Zhan W, Sismanidis P, Small C, doi:10.1016/j.enbuild.2017.03.066. Voogt J. 2019b. SUHI analysis using local climate zones- Fan H, Yu Z, Yang G, Yiu LT, Ying LT, Hung CH, Vejre H. 2019. How a comparison of 50 cities. Urban Clim. 28:100451. to cool hot-humid (Asian) cities with urban trees? An optimal doi:10.1016/j.uclim.2019.01.005. landscape size perspective. Agric For Meteorol. 265:338–348. Benton-short L, Keeley M, Rowland J. 2019. Green infrastructure, doi:10.1016/j.agrformet.2018.11.027. green space, and sustainable urbanism: geography’s impor- Fathi N, Bounoua L, Messouli M. 2019. A satellite assessment of tant role. Urban Geogr. 40(3):330–351. doi:10.1080/ the urban heat island in Morocco. Can J Remote Sens. 45 02723638.2017.1360105. (1):26–41. doi:10.1080/07038992.2019.1601007. Bowler DE, Buyung-Ali L, Knight TM, Pullin AS. 2010. Urban Feyisa GL, Dons K, Meilby H. 2014. Efficiency of parks in mitigating greening to cool towns and cities: a systematic review of urban heat island effect: an example from Addis Ababa. Landsc the empirical evidence. Landsc Urban Plan. 97(3):147–155. Urban Plan. 123:87–95. doi:10.1016/j.landurbplan.2013.12.008. doi:10.1016/j.landurbplan.2010.05.006. Filho WL, Icaza LE, Neht A, Klavins M, Morgan EA. 2018. Coping Brown RD, Vanos J, Kenny N, Lenzholzer S. 2015. Designing with the impacts of urban heat islands. A literature based urban parks that ameliorate the effects of climate study on understanding urban heat vulnerability and the change. Landsc Urban Plan. 138:118–131. doi:10.1016/j. need for resilience in cities in a global climate change landurbplan.2015.02.006. context. J Clean Prod. 171:1140–1149. doi:10.1016/j. Cao X, Onishi A, Chen J, Imura H. 2010. Quantifying the cool island jclepro.2017.10.086. intensity of urban parks using ASTER and IKONOS data. Landsc Geletič J, Lehnert M, Savić S, Milošević D. 2019. Inter-/intra-zonal Urban Plan. 96(4):224–231. doi:10.1016/j. seasonal variability of the surface urban heat island based on landurbplan.2010.03.008. local climate zones in three central European cities. Build Chavez JPS. 1996. Image-based atmospheric corrections- revisited Environ. 156:21–32. doi:10.1016/j.buildenv.2019.04.011. and improved. Photogramm Eng Remote Sens. 62 Gillner S, Vogt J, Tharang A, Dettmann S, Roloff A. 2015. Role of (9):1025–1036. street trees in mitigating effects of heat and drought at Choudhury D, Das K, Das A. 2018. Assessment of land use land highly sealed urban sites. Landsc Urban Plan. 143:33–42. cover changes and its impact on variations of land surface doi:10.1016/j.landurbplan.2015.06.005. temperature in Asansol-Durgapur development region. Gunawardena KR, Wells MJ, Kershaw T. 2017. Utilising green Egypt J Remote Sens Space Sci. 22(2):203–218. and bluespace to mitigate urban heat island intensity. INTERNATIONAL JOURNAL OF URBAN SUSTAINABLE DEVELOPMENT 263 Sci Total Environ. 584–585:1040–1055. doi:10.1016/j. Mahmoud SH, Gan TY. 2018. Long-term impact of rapid urbani- scitotenv.2017.01.158. zation on urban climate and human thermal comfort in Guo C, Schwarz N, Buchmann CM. 2018. Exploring the added hot-arid environment. Build Environ. 142:83–100. value of population distribution indicators for studies of doi:10.1016/j.buildenv.2018.06.007. European urban form. Appl Spat Anal Policy. 11(3):439–463. Marando F, Salvatori E, Sebastiani A, Fusaro L, Manes F. 2019. doi:10.1007/s12061-017-9225-7. Regulating ecosystem services and green infrastructure: Guo G, Wu Z, Chen Y. 2019. Complex mechanisms linking land assessment of urban heat island effect mitigation in the surface temperature to greenspace spatial patterns: evidence municipality of Rome, Italy. Ecol Modell. 392:92–102. from four southeastern Chinese cities. Sci Total Environ. doi:10.1016/j.ecolmodel.2018.11.011. 674:77–87. doi:10.1016/j.scitotenv.2019.03.402. Martilli A, Krayenhoff ES, Nazarian N. 2020. Is the urban heat Haq SMA. 2011. Urban green spaces and an integrative island intensity relevant for heat mitigation studies? Urban approach to sustainable environment. J Environ Prot. 2 Clim. 31:100541. doi:10.1016/j.uclim.2019.100541. (5):601–608. doi:10.4236/jep.2011.25069. Martos A, Pacheco-Torres R, Ordóñez J, Jadraque-Gago E. 2016. Hedquist BC, Brazel AJ. 2014. Seasonal variability of tem- Towards successful environmental performance of sustain- peratures and outdoor human comfort in Phoenix, able cities: intervening sectors. A review. Renew Sustain Arizona, U.S.A. Build Environ. 72:377–388. doi:10.1016/j. Energy Rev. 57:479–495. doi:10.1016/j.rser.2015.12.095. buildenv.2013.11.018. Moghbel M, Shamsipour AA. 2019. Spatiotemporal characteris- Jaganmohan M, Knapp S, Buchmann CM, Schwarz N. 2016. The tics of urban land surface temperature and UHI formation: bigger, the better? The influence of urban green space a case study of Tehran, Iran. Theor Appl Climatol. 137(3– design on cooling effects for residential areas. J Environ 4):2463–2476. doi:10.1007/s00704-018-2735-7. Qual. 45(1):134–145. doi:10.2134/jeq2015.01.0062. Morakinyo TE, Ouyang W, Lau KKL, Ren C, Ng E. 2020. Right tree, Jamei E, Rajagopalan P, Seyedmahmoudian M, Jamei Y. 2016. right place (urban canyon): tree species selection approach Review on the impact of urban geometry and pedestrian for optimum urban heat mitigation - development and level greening on outdoor thermal comfort. Renew Sustain evaluation. Sci Total Environ. 719:137461. doi:10.1016/j. Energy Rev. 54:1002–1017. doi:10.1016/j.rser.2015.10.104. scitotenv.2020.137461. Jamshidi S, Zand-Parsa S, Pakparvar M, Niyogi D. 2019. Nastran M, Kobal M, Eler K. 2019. Urban heat islands in relation Evaluation of evapotranspiration over a semiarid region to green land use in European cities. Urban For Urban Green. using multiresolution data sources. J Hydrometeorol. 20 37:33–41. doi:10.1016/j.ufug.2018.01.008. (5):947–964. doi:10.1175/JHM-D-18-0082.1. Ochola EM, Fakharizadehshirazi E, Adimo AO, Mukundi JB, Jenerette GD, Harlan SL, Buyantuev A, Stefanov WL, Declet- Wesonga JM, Sodoudi S. 2020. Inter-local climate zone differ - Barreto J, Ruddell BL, Myint SW, Kaplan S, Li X. 2016. Micro- entiation of land surface temperatures for management of scale urban surface temperatures are related to land-cover urban heat in Nairobi City, Kenya. Urban Clim. 31:100540. features and residential heat related health impacts in doi:10.1016/j.uclim.2019.100540. Phoenix, AZ USA. Landsc Ecol. 31(4):745–760. doi:10.1007/ Peng S, Piao S, Ciais P, Friedlingstein P, Ottle C, Bréon FM, Nan H, s10980-015-0284-3. Zhou L, Myneni RB. 2012. Surface urban heat island across Kim G, Coseo P. 2018. Urban park systems to support sustain- 419 global big cities. Environ Sci Technol. 46(2):696–703. ability: the role of urban park systems in hot arid urban doi:10.1021/es2030438. climates. Forests. 9(7):1–16. doi:10.3390/f9070439. Rogan J, Ziemer M, Martin D, Ratick S, Cuba N, DeLauer V. 2013. Knapp S, Jaganmohan M, Schwarz N. 2019. Climate regulation by The impact of tree cover loss on land surface temperature: diverse urban green spaces: risks and opportunities related to a case study of central Massachusetts using Landsat Thematic climate and land use change. In: Schröter M, Bonn A, Klotz S, Mapper thermal data. Appl Geogr. 45:49–57. doi:10.1016/j. Seppelt R, Baessler C, editors. Atlas of Ecosystem Services. apgeog.2013.07.004. Cham: Springer Nature Switzerland; p. 167–172. Rouhi H, Chamani N, Jafarnezhad J, Asgarian A. 2018. Knight T, Price S, Bowler D, King S. 2016. How effective is ‘green- Spatial assessment of the effects of in situ and neigh- ing’ of urban areas in reducing human exposure to ground- bourhood factors on urban land surface temperature level ozone concentrations, UV exposure and the ‘urban heat mitigation in a rapidly developing region. Int J Urban island effect’? A protocol to update a systematic review. Sustain Dev. 10(3):292–304. doi:10.1080/ Environ Evid. 5(1):1–6. doi:10.1186/s13750-016-0054-y. 19463138.2018.1522320. Krayenhoff ES, Voogt JA. 2016. Daytime thermal anisotropy of Saaroni H, Amorim JH, Hiemstra JA, Pearlmutter D. 2018. Urban urban neighbourhoods: morphological causation. Remote green infrastructure as a tool for urban heat mitigation: Sens. 8(2):108. doi:10.3390/rs8020108. survey of research methodologies and findings across differ - Lin W, Yu T, Chang X, Wu W, Zhang Y. 2015. Calculating cooling ent climatic regions. Urban Clim. 24:94–110. doi:10.1016/j. extents of green parks using remote sensing: method and uclim.2018.02.001. test. Landsc Urban Plan. 134:66–75. doi:10.1016/j. Senanayake IP, Welivitiya WDDP, Nadeeka PM. 2013. Remote landurbplan.2014.10.012. sensing based analysis of urban heat islands with vegetation Mabon L, Kondo K, Kanekiyo H, Hayabuchi Y, Yamaguchi A. cover in Colombo city, Sri Lanka using Landsat-7 ETM+ data. 2019. Fukuoka: adapting to climate change through urban Urban Clim. 5:19–35. doi:10.1016/j.uclim.2013.07.004. green space and the built environment? Cities. 93:273–285. Shashua-Bar L, Pearlmutter D, Erell E. 2009. The cooling effi - doi:10.1016/j.cities.2019.05.007. ciency of urban landscape strategies in a hot dry climate. 264 F. S. JAMALI ET AL. Landsc Urban Plan. 92(3–4):179–186. doi:10.1016/j. Voogt JA, Oke TR. 2003. Thermal remote sensing of urban landurbplan.2009.04.005. climates. Remote Sens Environ. 86(3):370–384. doi:10.1016/ Shiflett SA, Liang LL, Crum SM, Feyisa GL, Wang J, Jenerette GD. S0034-4257(03)00079-8. 2017. Variation in the urban vegetation, surface temperature, Wang Y, Bakker F, de Groot R, Wortche H, Leemans R. 2015. air temperature nexus. Sci Total Environ. 579:495–505. Effects of urban trees on local outdoor microclimate: doi:10.1016/j.scitotenv.2016.11.069. synthesizing field measurements by numerical Snir K, Pearlmutter D, Erell E. 2016. The moderating effect of modelling. Urban Ecosyst. 18(4):1305–1331. doi:10.1007/ water-efficient ground cover vegetation on pedestrian ther- s11252-015-0447-7. mal stress. Landsc Urban Plan. 152:1–12. doi:10.1016/j. Weng Q. 2009. Thermal infrared remote sensing for urban cli- landurbplan.2016.04.008. mate and environmental studies: methods, applications, and Sobrino JA, Jiménez-Muñoz JC, Paolini L. 2004. Land surface trends. ISPRS J Photogramm Remote Sens. 64(4):335–344. temperature retrieval from Landsat TM 5. Remote Sens doi:10.1016/j.isprsjprs.2009.03.007. Environ. 90(4):434–440. doi:10.1016/j.rse.2004.02.003. Weng Q, Lu D, Schubring J. 2004. Estimation of land surface Sodoudi S, Shahmohamadi P, Vollack K, Cubasch U, Che-Ani AI. temperature-vegetation abundance relationship for urban 2014. Mitigating the urban heat island effect in megacity heat island studies. Remote Sens Environ. 89(4):467–483. Tehran. Adv Meteorol. 2014:547974. doi:10.1155/2014/ doi:10.1016/j.rse.2003.11.005. 547974. Wheeler SM, Abunnasr Y, Dialesandro J, Assaf E, Agopian S, Statistical Center of Iran. 2020. [accessed 2020 Jan 10]. www. Gamberini VC. 2019. Mitigating urban heating in dryland amar.org.ir/english/Population-and-Housing-Censuses cities: a literature review. J Plan Lit. 34(4):434–446. Stewart ID, Oke TR. 2012. Local climate zones for urban tem- doi:10.1177/0885412219855779. perature studies. Bull Am Meteorol Soc. 93(12):1879–1900. Yan H, Wu F, Dong L. 2018. Influence of a large urban park on the doi:10.1175/BAMS-D-11-00019.1. local urban thermal environment. Sci Total Environ. Stewart ID, Oke TR, Krayenhoff ES. 2014. Evaluation of the ‘local 622–623:882–891. doi:10.1016/j.scitotenv.2017.11.327. climate zone’ scheme using temperature observations and Yang G, Yu Z, Jørgensen G, Vejre H. 2020. How can urban model simulations. Int J Climatol. 34(4):1062–1080. blue-green space be planned for climate adaption in doi:10.1002/joc.3746. high-latitude cities? A seasonal perspective. Sustain Cities Sun R, Chen A, Chen L, Lü Y. 2012. Cooling effects of wetlands in Soc. 53:101932. doi:10.1016/j.scs.2019.101932. an urban region: the case of Beijing. Ecol Indic. 20:57–64. Yu Z, Guo X, Zeng Y, Koga M, Vejre H. 2018. Variations in land doi:10.1016/j.ecolind.2012.02.006. surface temperature and cooling efficiency of green space in Tayyebi A, Shafizadeh-Moghadam H, Tayyebi AH. 2018. rapid urbanization: the case of Fuzhou city, China. Urban For Analyzing long-term spatio-temporal patterns of land surface Urban Green. 29:113–121. doi:10.1016/j.ufug.2017.11.008. temperature in response to rapid urbanization in the Zhang Y, Murray AT, Turner BL. 2017. Optimizing green space mega-city of Tehran. Land Use Policy. 71:459–469. locations to reduce daytime and nighttime urban heat island doi:10.1016/j.landusepol.2017.11.023. effects in Phoenix, Arizona. Landsc Urban Plan. 165:162–171. Tran DX, Pla F, Latorre-Carmona P, Myint SW, Caetano M, doi:10.1016/j.landurbplan.2017.04.009. Kieu HV. 2017. Characterizing the relationship between land Zhou X, Okaze T, Ren C, Cai M, Ishida Y, Watanabe H, Mochida A. use land cover change and land surface temperature. ISPRS 2020. Evaluation of urban heat islands using local climate J Photogramm Remote Sens. 124:119–132. doi:10.1016/j. zones and the influence of sea-land breeze. Sustain Cities isprsjprs.2017.01.001. Soc. 55:102060. doi:10.1016/j.scs.2020.102060.
Journal
International Journal of Urban Sustainable Development
– Taylor & Francis
Published: May 4, 2021
Keywords: Urban parks; land surface temperature; local climate zones; semi-arid; sustainability
