Environmental variables and air quality: implications for planning and public health

1.1 Tall buildings and the city

Most of the world’s population currently lives in urban areas. Trends in urban population growth have resulted in the expansion of urban land coverage, a variable known to affect the urban climate (IPCC 2021), resulting in thermal discomfort and deteriorated air quality. Poor air quality is a direct result of transportation intensification and urban activity in combination with atmospheric conditions that critically influence the urban microclimate and, in turn, the health and wellbeing of urbanites (Grimm et al. 2008; Levi & Barnett-Itzhaki 2021).

Identifying the above issues led to the commencement of climate studies correlating climate change, especially the urban heat island (UHI) effect, to population growth and urbanisation’s characteristics, among them increasing traffic and energy use. According to Angel et al. (2001), the implications of urban growth for climate change are complex when viewed at local, regional and global levels. The densification of cities is currently considered by many as an efficient way to curb urban sprawl and anthropogenic impact (Glaeser 2011; Mills 2007), even though cities are direct and major consumers of energy, as well as being significant air polluters (Mage et al. 1996; Molina & Molina 2004). Cities also indirectly contribute to global climate change due to what the Intergovernmental Panel on Climate Change (IPCC) defined as their ‘insatiable appetite’ for energy and materials, associated with the production of waste and air pollution (IPCC 2019).

The enhanced urbanisation and ensuing crowded areas create environmental degradation and contingencies (Evance 1983), aggravating microclimate and thermal discomfort, as well as deteriorating air quality. The current high levels of air pollution that cities portray today have considerable consequences to the health of humans, as well the various urban ecosystems (EEA 2019).

The adoption of in-situ monitoring methodologies and systems has increased the accuracy needed to portray the urban environment and to understand the complexity—physical, environmental, social, economic, etc.—that governs the planning, design and construction activities within the urban fabric. Of special interest is the inclusion of tall building design and spatial distribution. The Council of Tall Buildings and Urban Habitat (CTBUH) defines a building of 14 or more floors—or more than 50 m in height—as being within the threshold of the ‘tall building’ definition (Choi et al. 2017; Lotfabadi 2015). The terms ‘tall’, ‘high-rise’ and ‘skyscraper’ are used here interchangeably, albeit all of them meaning buildings defined by the CTBUH as ‘tall’ and above.

When studying the effects of tall buildings within the urban fabric, the height and arrangement of neighbouring buildings must take into consideration urban wind direction and velocities, and solar radiation and/or shade, which may influence a cluster’s microclimate, thermal behaviour, outdoor as well as indoor thermal comfort levels (Saroglou et al. 2019), as well as the effect tall buildings have on urban wind patterns (Saroglou et al. 2022) and pollution dispersion.

In that respect, high-rise buildings near busy roads can have both positive and negative effects on air quality in their vicinity. It is often assumed that they can act as physical barriers that partially block or divert pollution from reaching certain areas and aid in dispersing pollutants vertically. However, high-rise buildings along roads can also create a ‘canyon effect’ where air flow is restricted and pollutants are trapped within the space between buildings. This can lead to higher concentrations of pollutants in these areas, e.g. particulate matter (PM), nitrogen oxides (NO_x) and volatile organic compounds (VOCs), which can have adverse health effects. The presence of high-rise buildings can create localised microclimates where air quality differs from surrounding areas. Factors, e.g. wind patterns, building orientation and surrounding structures, can all influence pollutant distribution patterns.

Living in high-rise buildings can have different health implications. The creation of pedestrian-friendly environments may encourage walking and physical activity. Conversely, reliance on elevators and lack of nearby usable green spaces can discourage physical activity. Additionally, high-density living in urban areas with high-rise buildings may increase exposure to air pollution and noise, potentially impacting respiratory health and overall wellbeing.

Factors, e.g. limited access to nature, social isolation and lack of community spaces, within high-rise developments can contribute to feelings of loneliness, stress and depression. On the other hand, well-designed high-rise developments with amenities, e.g. communal areas, green spaces and recreational facilities, may promote social interaction and a sense of belonging, positively impacting mental health.

The presence of high-rise buildings can reflect broader social and economic dynamics within a city. In some cases, high-rise developments may exacerbate socio-economic inequalities by gentrifying neighbourhoods and displacing lower income residents. Access to affordable housing, essential services and amenities within high-rise developments can influence health outcomes and contribute to social equity within urban communities. The construction and operation of high-rise buildings have environmental implications, including energy consumption, resource use and waste generation. Sustainable design principles, e.g. energy-efficient building technologies, green building materials and renewable energy sources, can minimise the environmental footprint of high-rise developments and promote healthier urban environments.

Considered from this perspective, the aim of the current research work that forms part of a broader research effort is to study and understand the interrelations of tall buildings and their urban context, attempting to build a theoretical and empirical base aimed at enhancing the sustainability of this typology (Saroglou et al. 2019, 2022).

1.2 Air pollutants within the urban environment

Research on air pollution has indicated lower concentrations of pollutants in urban parks (Paoletti et al. 2011) compared with adjacent urban canyons (Kuttler & Strassburger 1999; Yin et al. 2007; Cavanagh et al. 2009; Cohen et al. 2012). Air pollution was measured according to the size and amount of PM (also called particle pollution), i.e. the mixture of solid particles and liquid droplets found in the air. Some particles, e.g. dust, dirt, soot or smoke, are large or dark enough to be visible to the naked eye. Others are so small they can only be detected using a scanning electron microscope. Fine particles < 2.5 μm in diameter (PM2.5) have high residence time in the air and are both the leading cause of reduced visibility (haze) in big cities, rural areas, national parks and wilderness areas in many countries, as well as being inhalable and causing respiratory-related illnesses. Most particles form in response to complex reactions of chemicals, e.g. sulfur dioxide (SO₂) and NO_x, pollutants emitted from power plants, industry, combustion heaters (including household ones) and automobiles.

The size of the particles is linked to their potential for causing health problems, raising morbidity and mortality (Fewell et al. 2007; Eftim et al. 2008; Bell et al. 2015). Small particles of inhalable size, < 2.5 μm in diameter, pose the most significant issues because they can penetrate deep into the lungs; some may even enter the bloodstream. When discussing the continuum of effects from PM2.5 exposure, research should explain how this continuum is impacted by the various concentrations at which they have been observed. For instance, exposure to ambient air pollution worldwide is related to 4.2 million premature deaths annually (WHO 2020). These include ambient air pollutants with a PM diameter < 10 μm (PM10) and ambient fine pollutants of PM < 2.5 μm (PM2.5), NO_x, SO₂ and carbon monoxide (CO). The related negative health outcomes include respiratory diseases, e.g. asthma, chronic obstructive pulmonary disease (COPD) and pneumonia, and cardiovascular related morbidity, e.g. enhanced thrombosis, elevated arterial blood pressure and enhanced arteriosclerosis/atherosclerosis (Khafaie et al. 2016; WHO 2013). Recent studies have shown an association with increased COVID-19 morbidity and mortality (De Angelis et al. 2021; Levi & Barnett-Itzhaki 2021).

1.3 Tel aviv’s urban climate and air quality

A growing body of publications stemming from research work have investigated the rapid changes of the Tel Aviv urban environment. Using Tel Aviv as a case study of fast-growing cities, Alpert et al. (2019) have shown a clear correlation between urbanisation and increased precipitation downwind of the urban area, as well as an upwind decrease over the coastal region. These findings are well-correlated with the expansion of the urban area and rainfall urban anomalies, which were found to be in the order of 50–100 mm/yr. Saaroni et al. (2000) have identified a significant UHI on a stable winter’s day. Specifically, tall buildings of 20 floors located in the study area were shown to be the warmest buildings, as opposed to a small residential housing area which was documented to be relatively cold. At night, within a radius of 5 km, air temperature differences were registered between the city centre areas and the south-eastern margins of the city, reaching up to 6°C at street level and about 2°C at roof level. Their study showed thermal variations on a microscale level, caused by coverage type, land use, building materials and activity patterns, with tall buildings posing an urban microclimate challenging prototype. Other publications stress the complexity of densifying urban centres with new buildings reaching upward. Gadish et al. (2023) stressed the interdependence between all the various climatic variables that impinge on thermal stress. They state that increasing building height can moderate thermal stress due to deeper shading, but this benefit is dampened as the air becomes warmer and more stagnant, and the addition of anthropogenic heat emissions further clouds the picture.

Those are significant results regarding the interactions between urbanisation and urban climate, not least since Tel Aviv is implementing a skyward development plan which is changing the surface roughness. Mandelmilch et al. (2020) examined the effect of urban spatial patterns on heat exposure in the city of Tel Aviv. They showed that at midday the city is ~3.6°C warmer than the surrounding rural area 2 m above ground. They also found that parks had a cooling effect only during the hot hours of the day (9:00–17:00 hours), yet lack of urban vegetation in the southern part of the city caused land surface temperature there to be hotter by ~7–9°C compared with the northern city part. Compact midrise clusters were found to be hot spots, whereas compact low-rise clusters were less heat vulnerable. Such conclusions have potentially significant projections regarding design guidelines toward UHI mitigation and urban environment improvement.

Cohen et al. (2014) investigated the impact of an urban park on air pollution and noise levels, concluding that the park has a mitigating effect on both, albeit with several reservations regarding the park’s mitigation efficiency for specific pollutants, e.g. CO and ozone (O₃). However, a continuum of green residential neighbourhoods and urban parks (as opposed to isolated green spots, especially such located along major roads or parking areas) has also been shown to have a significant mitigating effect on UHI (Rotem-Mindali et al. 2015).

The projections of all such data become all the more pressing when the rising morbidity and mortality attributed to environmental degradation are considered. In a recently published report, the Israel Union for Environmental Defense (IUED, Adam Teva V’Din—Man Nature and Law) summarised recent reports by the Ministry of Environmental Protection and the Ministry of Health stating that, in 2023, air pollution contributed to 5510 premature deaths, surpassing by far previous estimates. The vast majority of these deaths were linked to exposure to fine particulate matter (PM2.5), with additional cases attributed to nitrogen dioxide (NO₂) and O₃ exposure. Concentration of PM2.5 in Israel is more than three times the World Health Organization’s (WHO) annual air quality guidelines (IUED 2025).

The research module presented in this paper examines the effect of urban geometry on air pollutant concentrations during the summer season in a hot Mediterranean climate and analyses the influence of high-rise buildings on the concentration of air pollutants within the urban fabric.

2.1 Study site

The study was conducted in Tel Aviv, Israel, situated on the eastern coast of the Mediterranean Sea (32°060 N, 34°470 E). The city’s climate is defined as subtropical Mediterranean (Bitan & Rubin 1994), classified as Csa in the Köppen–Geiger classification. Climate conditions, stable during the summer, are characterised by hot temperatures, clear skies, western and north-western winds at moderate velocity (0–1.9 m/s), and solar radiation daily maximum average > 3.07 kWh/m². UHIs range between 0.8 and 2.2°C at night in summer (Bitan et al. 1992) and between 2 and 5°C in winter (Saaroni et al. 2000).

Tel Aviv (population of 474,530 in 2022) is the core of Israel’s largest metropolitan area (out of a total population of 9.757 million) (CBS 2022). Studies have shown that Tel Aviv’s metropolitan area’s accelerating growth has driven the creation of UHIs (Bitan et al. 1992; Saaroni et al. 2000) as well as high concentrations of air pollution, both being signs of intense transportation and energy consumption (Cohen et al. 2014). Results of a study conducted there predicted a growing and high correlation between population and local warming (Itzhak-Ben-Shalom et al. 2016).

To examine the impact of urban design on the urban climate and air pollution, three representative types of urban areas were selected for monitoring (west to east, upwind to downwind). They were selected in close proximity to one another and at a roughly similar distance from the sea; for a more accurate comparison between differences in their microclimatic conditions, see Figures 1 and 2 (green markers). The urban sites are as follows:

• a pedestrian bridge (Yehudit Bridge) crossing an urban highway (Ayalon highway) and railway¹

• a recently built office high-rise cluster of two 42-storey towers, Alon 1 and 2² and

• a low, two-to-three stories residential area (Bitzaron).

All three monitoring sites are located in the eastern part of the city, downwind of the prevalent western wind in Tel Aviv known as the ‘Mediterranean breeze’. An initial assumption was that being downwind from one of the major traffic corridors of the region, i.e. the urban highway and railway, and the prevailing western winds, meant that pollution concentration would be higher from other parts of the city. Another assumption was that the high-rise cluster would induce a Venturi effect between the towers, affecting pollutant concentration.

Figure 1 shows the location of the three in-situ monitoring sites (green markers) alongside six of Tel Aviv’s permanent meteorological and air quality stations (red markers). These are operated by the Ministry of Environmental Protection and the Ministry of Agriculture and Rural Development. Data obtained from those permanent stations, located in close proximity to the in-situ monitoring stations, were used to validate the study’s monitoring results. Several of these meteorological stations collect site-specific data, e.g. dominant air pollution particles found in the area.

Figure 1

The HaShalom and HaHagana train station monitoring systems collect temperature, air quality and relative humidity data. The Kremenetzky residential area station collects temperature and wind velocity data. The City Center station monitors wind speed, whereas the Lehi and Levinsky stations, also located within residential areas, monitor air quality.

Figure 2 shows in greater detail the urban area under study and the locations of the three in-situ monitoring stations. The studies included the collection of air pollution data, as well as environmental variables, e.g. air temperature, relative humidity, wind velocity and direction.

Figure 2

2.2 Study period

This study was conducted over three days (Thursday, Saturday and Sunday, adjusted to the local work week) in July 2021, in the midst of Tel Aviv’s summer season. Data were monitored at an estimated average interval of 30 min, between 08:00 and 20:00 hours daily. In most of the world, the working week extends from Monday to Friday, with the weekend on Saturday and Sunday. However, in Israel, the working week starts on Sunday and continues through Thursday, with the weekend being on Friday and Saturday. This means that the study covered two full working days (Thursday, 1 July and Sunday, 4 July). Saturday, 3 July completes the weekend. July was chosen because it is considered among the hottest months of the year, with solar radiation at its height, peaking at a daily maximum average of > 3.07 kWh/m².

2.3 In-situ measurements

Air pollution was monitored as follows: NO_x was monitored by Monitor Labs ML 9841A NO NO2 NOx (EU); PM2.5 (μg/m³) was monitored by a Continuous Ambient Particulate TEOM™ Monitor 1400ab (EU), and Dylos PM1700 (USA). Data were collected consecutively and stored in data loggers. Climatic variables (air temperature, relative humidity, wind direction and velocity) were monitored by HOBO USB Microstations H-21 during the same period.

All instruments were calibrated, checked and compared under the same conditions. A total of two monitoring stations per location was used to calibrate the data between them. Figure 3 depicts the Alon Towers location with the in-situ installations.

Figure 3

3.1 Air temperature (°c)

Data were collected for the three consecutive days investigated (over 12-h periods, 08:00–20:00) during July, one of Israel’s hottest summer months, at the three representative types of urban spaces selected: Yehudit Bridge, Alon 1 and 2 towers, and Bitzaron. All three sites are within a ~250 m radius. Collected data were compared with those of the three adjacent urban monitoring stations (HaShalom, HaHagana, Kremenetzky) (Figures 2, 4). The air temperatures registered at the meteorological stations at the HaHagana and HaShalom stations are taken as reference points for the Yehudit bridge area; the Kremenetzky station temperatures are taken as the reference point for Bitzaron residential area.

Figure 4 depicts the results of air temperature comparisons between the three sites and the three permanent urban meteorological stations, for Thursday, 1 July. Results indicate that the lowest temperatures during early morning (08:00–10:00) were observed at Yehudit Bridge, HaShalom station, HaHagana station and Kremenetzky station. The highest temperatures during daytime hours (14:00–17:00) were measured in Bitzaron residential area. Differences were more pronounced at HaShalom station (due to its location under the highway). The highest temperatures were measured at midday, when the maximum temperature in the residential area ranged between 33 and 36°C, in comparison with between 29 and 31°C at Kremenetzky station and Yehudit Bridge, and in close relation to the temperatures at the base of Alon Towers.

Figure 4

Figure 5 delineates the differences in the air temperatures (°C), for Saturday, 3 July. The highest temperatures for the three monitored sites were measured from 08:00 to 18:00 hours in the residential area, reaching 36°C between 13:00 and 14:00, and dropping considerably to match the temperatures of the other locations after 17:00. During the same period, temperatures measured at Yehudit Bridge and HaShalom station ranged between 31 and 32°C, until about 17:00 hours, when they started to drop. The two stations, HaShalom and Yehudit Bridge, presented very similar data on Saturday as opposed to Thursday and Sunday. This is attributed here to the railway not operating during the weekend, and the moderate traffic on Saturday in Israel.

Figure 5

Figure 6 portrays the differences in air temperature for Sunday, 4 July. Data show a similar trend with Thursday, 1 July (Figure 2). During the day, lowest temperatures were noted, again, during the early morning hours from 8:00 to 10:00 for Alon Towers and Yehudit Bridge, and for HaHagana and Kremenetzky stations. Highest temperatures were observed in the residential area throughout the day, with maxima measured between 11:00 and 18:00. Temperatures were more pronounced at midday, with maximal temperature in the residential area being ~31–35°C compared with the ~28–30°C at the other four sites. Temperatures in Alon Towers were initially the lowest and increased almost to match those at the bridge and HaHagana station from 15:00 to 19:00 hours, when they dropped again.

Figure 6

3.2 Wind velocity

Figures 7, 9 depict the timeline for wind and gust velocities at the three study sites. The City Center and Kremenetzky monitoring stations serve as reference points. For 1 July, according to the data depicted in Figure 7, at midday the highest difference between the Kremenetzky weather station and Bitzaron residential area is ~5 m/s. The lowest wind velocities measured are in the evening hours, when the difference between the two points is < ~2.5 m/s. There is a correlation between the high temperatures in the residential area on Thursday (Figure 4) and the low wind velocities (Figure 7), suggesting an overheating of the urban environment.

Figure 7

According to Figure 7, wind velocity comparisons between the three sites under study and the three permanent urban meteorological stations showed that, during daytime, the highest wind speeds were measured in Kremenetzky. This may well be due to the upwind urban fabric being relatively unobstructed which, though, may very soon change, as an adjacent 185 m-long urban block is under construction which may potentially block wind circulation and ventilation. A similar phenomenon is already happening in Bitzaron residential area, downwind from Alon Towers, and several other adjacent areas.

Figure 8 depicts the wind velocities for Sunday, 3 July. Higher wind velocities are measured at the Alon Towers and Kremenetzky and City Center weather stations at midday and remain higher in comparison with the other stations throughout the day. Yehudit Bridge, on the other hand, portrays wind velocities of ~2.8 and 3 m/s lower, while Bitzaron residential area presents very low wind velocities throughout the day (0.5–1.0 m/s). However, towards the evening, after 19:00 hours, wind velocities drop considerably in all stations, with Bitzaron residential area having almost no wind circulation.

Figure 8

Figure 9 depicts wind velocity results for Sunday, 4 July. Wind velocity ranges between the Alon Towers and City Center and Kremenetzky stations were between 4 and 5 m/s, with maximum differences of ~1 m/s between them for most of the day. However, after 17:00 hours, differences between these three stations became negligible, and after 19:00 started to drop even further. Wind speed comparisons between Alon Towers and Yehudit Bridge show maximum differences of ~2.8 m/s between 10:00 and 12:00 hours, and < 1 m/s after 17:00.

Figure 9

3.3 Air pollution monitoring

In this section the results of air temperature and wind velocity are associated with the results of air pollution. Figure 10 depicts data for Thursday, 1 July, for Bitzaron residential area. Although PM2.5 concentrations were above the threshold considered healthy for sensitive groups during early morning hours (30 μg/m³), after 10:00 hours they decreased to moderate values (17.5–12.5 μg/m³) and became stable for the rest of the day. The wind velocity during the same early morning hours was close to zero, with gusts at 1 m/s, while after 10:00 hours, wind velocity rose to 0.5–1 m/s and gust speed rose to 2.0–2.5 m/s, and remained stable until calm weather set in after 19:00. Results indicate that lower PM2.5 concentrations correlate with higher wind and gust values.

Figure 10

Figure 11 shows that PM2.5 concentrations between the Alon Towers were considerably above the healthy level threshold for sensitive groups (55–35 μg/m³). However, after 10:00 hours, PM2.5 concentration dropped to moderate levels (35.5–20.0 μg/m³). Stable PM2.5 values were observed during the rest of the day, with a sharp peak at 18:00 hours. Wind velocity at those same early morning hours was 1.5 m/s, and gust speed was 3.5 m/s. After 10:00 hours, wind velocity rose to 3.5–4.0 m/s and gusts to 6.0–6.5 m/s. The wind and gust speed observed increases are associated with corresponding PM2.5 concentration reductions, suggesting that enhanced air movement contributes to the dispersion of fine PM in the urban environment.

Figure 11

Figure 12 shows that on Thursday, 1 July, during the morning hours at Yehudit Bridge over Ayalon highway and railway, PM2.5 concentrations were above the healthy threshold for sensitive groups (55–35 μg/m³). After 10:00 hours, concentrations dropped to a healthier moderate level (~25 μg/m³), remaining relatively stable for the rest of the day, peaking slightly at 32 μg/m³ around 16:00 hours before declining to 15 μg/m³. During the early morning hours, wind velocity measured approximately 1.5 m/s, with gusts reaching 3.5 m/s. After 10:00 hours, wind speeds increased to 2.0–2.5 m/s, with gusts rising to 5–6 m/s, likely contributing to the dispersion of PM2.5. However, wind and gust velocities decreased again during the afternoon, returning to around 1.5 and 2.5–3.0 m/s, respectively. These patterns suggest a correlation between increased air movement and reduced PM2.5 concentrations. Observed gust accelerations can be explained by the Bernoulli principle and the Venturi effect, whereby airflow is amplified between low- and high-rise buildings, enhancing localised ventilation and pollutant dispersion.

Figure 12

Figure 13 presents Bitzaron residential area data for Saturday, 3 July. During the morning hours, PM2.5 concentrations were within the moderate range (30 μg/m³). After 10:00 hours, levels briefly decreased to 25 μg/m³, before rising again to 30 μg/m³ by 20:00 hours. Wind velocities in the early morning were very stagnant, with speeds close to 0 m/s, and gusts around 1 m/s. After 10:00 hours, wind velocities slightly increased to 0.5–1.0 m/s, and gusts reached 1.5–2.5 m/s, remaining relatively steady throughout the day, before dropping in the evening. Air temperatures during this time fluctuated between 30 and 34°C. The persistently low wind speeds and modest gust levels may have contributed to PM2.5 concentration limited dispersion throughout the day.

Figure 13

Figure 14 illustrates conditions at Alon Towers for Saturday, 3 July. During the morning hours, PM2.5 concentrations were moderate (28–30 μg/m³). However, after 09:30 hours, concentrations gradually rose to levels considered unhealthy for sensitive groups (35.5–55.4 μg/m³). In the early morning, wind velocity was approximately 2 m/s, with gusts reaching 4 m/s. After 09:30 hours, wind speeds rose to 4.0–4.5 m/s, with gusts picking up to 6.5–7.0 m/s, before gradually dropping by 19:00 hours to 1.5 and 2.7 m/s, respectively. Throughout the day, air temperature remained stable, ranging between 29 and 30°C. Despite stronger winds after 09:30 hours, PM2.5 concentrations continued to rise, indicating that local emissions may have hindered pollutant dispersion.

Figure 14

Figure 15 shows the conditions at Yehudit Bridge, for Saturday, 3 July. During the morning hours, PM2.5 levels reached 35 μg/m³, indicating unhealthy conditions for sensitive populations. After 10:00 hours, concentrations rose sharply, peaking at 55.5 μg/m³ by 12:00, remaining elevated in the unhealthily range (55–60 μg/m³) throughout the afternoon. Early morning wind velocities were 2.0–2.5 m/s, with gusts between 4.5 and 5.5 m/s. However, after 11:00 hours, both wind and gusts dropped steadily to 1.5 and 2.5 m/s, respectively. Air temperature remained stable at 31–32°C throughout the day.

Figure 15

According to regional weather reports, a heavy and prolonged heatwave, originating in North America, moved towards the Mediterranean region (Zachariah et al. 2023). This hot air mass became trapped under strong barometric pressure systems at both ground levels and in the upper atmosphere. As temperature rose, an increase in the concentration of solid particles was also observed. These weather conditions likely sustained the high PM2.5 levels on Saturday, 3 July, by limiting air mixing and slowing dispersion, leading to particle build up in the urban canyon.

Figure 16 portrays Bitzaron residential area conditions for Sunday, 4 July. Morning PM2.5 concentrations were moderate (33 μg/m³), likely due to residual pollution from the previous day. Levels gradually declined to 13–15 μg/m³ by 14:00 hours, before rising again to 25 μg/m³ by 20:00. Early morning wind velocity hours was close to zero, with gusts of 1.5 m/s. After 10:00 hours, wind speeds increased slightly to 1 m/s and gusts to 2.5–3.5 m/s, but both weakened again by 20:00 hours, reaching 0 and 1 m/s, respectively. Air temperature rose from 30°C in the morning to a peak of 34°C at 16:00 hours, before gradually cooling to 27°C by 20:00.

Figure 16

Figure 17 shows environmental conditions at Alon Towers for Sunday, 4 July. In the morning, residual PM2.5 concentrations from the previous day were moderate (34 μg/m³). After 10:30 hours, levels gradually declined to ~19.5 μg/m³. Early morning wind speeds were 2.5 m/s, with gusts at 5 m/s. After 09:30 hours, wind velocity increased to 4–5 m/s and gusts rose to 6.5–7.5 m/s, both decreasing by 20:00 hours to 2.5 and 4.5 m/s, respectively. Temperature remained stable throughout the day at 29–30°C.

Figure 17

Figure 18 illustrates the environmental conditions at Yehudit Bridge, for Sunday, 4 July. PM2.5 concentrations were moderate during morning (29–34 μg/m³), gradually declining after 11:30 hours to 17 μg/m³ by 18:00, before rising to 25 μg/m³ by 20:00. Wind velocities started at 2.2 m/s, with gusts of 5 m/s, and then dropped to 1.5 and 2.5 m/s, respectively. Temperatures remained consistently high at 31–32°C. Elevated PM levels may be linked to the region’s prolonged heatwave, which likely limited atmospheric mixing and pollutant dispersion.

Figure 18

3.4 The four train stations

Data from the four permanent monitoring sites found at railway stations located throughout the city are presented. HaShalom station’s data were chosen in order to present a broader perspective regarding public (HaShalom and HaHagana stations) and private (Lehi and Levinsky stations) transportation impacts on the level of air pollution.

Figure 19 presents PM2.5 concentrations for all stations for Thursday, 1 July. At the Lehi and Levinsky stations a stable, relatively low pollution level exists. This represents the presence/effect of private transportation but not the axis created by railway lines. At 11:00 hours, during heavy traffic, a significant decline of pollution at the two stations is noted, which continued until 20:00. In contrast, the HaShalom and HaHagana stations had high PM2.5 pollution levels (35 μg/m³), with a sharp rise at 15:00 hours, especially at the HaShalom flyover, with another increase in the afternoon. These trends can be associated with the hours of heavy public transportation traffic (Figure 20).

Figure 19

Figure 20

On Saturday, 4 July, the official Jewish day of rest, public transportation is unavailable. And yet, on Saturdays, Israel is characterised by a significantly low use of private means of transportation. This observation is supported by the data collected at HaShalom station, located in a dugout corridor below the level of Ayalon highway, where variations were found in the data which differed from those collected at the other sites selected. At the Lehi and HaHagana stations, at 12:00 and 16:00 hours on Saturday, observed changes were directly influenced by the concurrent climatic conditions at the station. This may be easily illustrated by the high level of air pollution registered at the station, which can be related to the station’s low topographical position. At the Lehi and Levinsky sites, pollution reached above normal, but not extreme levels (≤ 30 μg/m³), indicating higher pollution in sensitive areas. Hence, given the absence of vehicular traffic, there is no circulation to strengthen the pollution measured at those sites. Moreover, Figure 21 indicates that the wind and gust velocities were low at the time, and on the decline.

Figure 21

At the beginning of Sunday’s (4 July) measurements, HaHagana and HaShalom stations start at a PM2.5 level of 45 μg/m³, whereas Levinsky and Lehi start lower, at more acceptable levels. This trend continued for the entire day, from 11:00 to about 20:00 hours. At HaHagana station, the situation remained stable, with readings decreasing to 40 μg/m³ even after 11:00. However, at HaShalom station, the pollution level returned to 60 μg/m³ at that same hour. After 12:00–13:00 hours, a decline at HaShalom station was observed, stabilising at the level observed at HaHagana station during those same hours. Throughout the day, normal values were registered, declining in tandem with the setting sun and the reduced traffic.

To conclude, the pollution values registered showed that the location of the HaShalom meteorological station, below ground level and partly enclosed by the train platform, may explain why it obtained the highest PM2.5 values on each measurement day during the entire period of fieldwork. The HaHagana meteorological station is also located on a railway station platform, hence the high values here, too. As the Lehi and Levinsky meteorological stations are located relatively far from major roads and railway stations, lower values were registered there.

4.1 Air temperature

Results illustrate the heterogeneous thermal behaviour of adjacent urban typologies during the summer season in Tel Aviv, collected over three consecutive days. Data reveal notable spatial and temporal variations in microclimatic conditions across different urban settings. Consistently, the residential area exhibited the highest daytime air temperatures, peeking between 33 and 36°C, particularly during midday hours (13:00–16:00). In contrast, Yehudit Bridge and Alon Towers demonstrated lower temperatures, ranging between 28 and 32°C. These sites benefit from the shading and airflow conditions created by the elevated structures and less dense surrounding urban fabric.

Saturday’s reduced traffic and rail inactivity significantly influenced the microclimate at Yehudit Bridge, aligning its temperature profile with that of HaShalom station. On the other hand, Alon Towers exhibited a distinct behaviour. Initially registering the lowest temperatures and then aligning more closely with Yehudit Bridge by mid-afternoon. The pattern suggests a delayed heat absorption probably linked to shading, as well as the higher wind velocities between the towers, as discussed below.

Findings affirm that low-rise residential zones are more susceptible to heat accumulation, while high-rise commercial zones and infrastructure sections, such as bridges, can mitigate peak air temperatures, especially during the first part of the day. Such microclimatic differences are crucial for urban planning and climate adaptation strategies, which could also include increased vegetation and smart materials, aiming to reduce urban heat stress and improve thermal comfort during extreme heat events.

4.2 Wind velocity

Across all three study days, wind speeds were consistently lower in Bitzaron residential area, ranging between 0.5 and 1.0 m/s. The reduced ventilation is likely linked to the dense low-rise configuration of the neighborhood, and its downwind position relative to Alon Towers and other high-rise development which obstruct natural airflow. Findings indicate a consistent urban overheating effect.

Saturday, 3 July, data demonstrated higher wind velocities at the Alon Towers, Kremenetzky and City Center stations due to their relatively unobstructed upwind orientation, while Yehudit Bridge and Bitzaron remained relatively stagnant. By evening, all locations portrayed low velocities, reinforcing the area’s poor air quality and thermal discomfort.

Sunday, 4 July, data had the lowest overall wind velocities of the study period. Although small differences were present between all stations, the Bitzaron area was again the most vulnerable, while also Yehudit Bridge recorded moderately low wind speeds. Despite Sunday morning heavy traffic, there was no significant traffic-induced influence on wind velocities, suggesting the urban fabric shape affects airflow more consistently than temporary human activities.

These results highlight a clear link between urban form and wind circulation, with low-rise and dense areas being significantly disadvantaged in terms of natural ventilation. Results also demonstrate the importance of ground-level environmental monitoring and climate-sensitive urban design strategies, in particular when new development projects may compromise natural ventilation and intensify heat-related stress.

4.3 Air pollution

Measurements over a three-day period in early July 2021, during an intense regional heatwave, revealed key insights into how wind velocity, temperature and local morphology influence air quality in dense urban settings.

In Bitzaron residential area, low wind speeds and limited air circulation resulted in stable but elevated PM2.5 concentrations. On days with reduced anthropogenic activity, e.g. Saturday, 3 July, air pollution levels reflected broader meteorological events rather than localised emissions. A small increase in wind velocity, in the range of 1 m/s, was shown to have meaningful pollution clearing effects. This suggests that even minor enhancements in urban ventilation can contribute to improved air quality at the neighborhood level.

At Alon Towers, PM2.5 concentrations remained high throughout the day despite fluctuations in wind velocity. Although early morning gusts were > 6 m/s, pollutant levels continued to rise, indicating that the local high-rise canyon geometry may trap pollutants. Therefore, the persistent high concentrations, even during periods of high wind velocities, underscore the influence of urban morphology over wind-driven dispersion alone, meaning that the canyon effect further intensified air stagnation and pollutant retention.

Yehudit Bridge presented patterns similar to Alon Towers, with PM2.5 levels peaking at midday despite moderate wind and gust activity. Despite moderate wind activity, pollution persisted, and on Saturday, 3 July, remained elevated, peaking at a hazardous 60–65 μg/m³ throughout the day. This pattern points to the influence of other factors, i.e. regional haze, high temperatures and limited vertical mixing, likely linked to the intense heatwave. These meteorological conditions hindered pollutant dispersion and allowed solid particles to accumulate along the bridge’s corridor and nearby canyons. On Sunday, 4 July, a notable drop was observed in pollutant accumulation probably attributed to the train and traffic activity resuming. Although both contribute to pollution emissions, their movement appears to promote wind circulation, supporting pollutant dispersion.