Environmental, Economic, and Health-Related Impacts of Increasing Urban Greenery Cover ^†

Abstract

An integrated approach, including statistical data elaborations and microclimate simulations, was presented in this paper to assess the impact of increasing the urban greenery cover in two communities in Ontario on the urban environment, air quality levels, health, and economic responses. The study also aimed to prove the association between ambient temperature and air quality. The correlations between meteorological parameters and air pollutants showed that the ozone and fine particulate matter daily mean concentrations are positively correlated with the mean temperature. The increase in the urban greenery cover confirmed a reduction in mean air temperature of 2 °C and daily average energy savings of 0.16 kWh/m². With the linkage to other responses, the results demonstrated a potential enhancement in all-cause mortalities and economic benefits.

Published: 25 May 2022

1. Introduction

The combination of climate change, Urban Heat Island (UHI), and heatwave events leads to higher daytime temperatures, causes excessive heat stress for urban dwellers, and increases heat-related mortality [1,2]. Meanwhile, the consequences of the UHI and the frequency and duration of heatwaves around the world are becoming more evident [1,3,4]. The Canadian Environment Health Atlas (CEHA) and the Toronto Public Health Department estimated that 120 people die in the Greater Toronto Area (GTA) annually because of high temperatures. The predictions indicate that heat-related mortality will be doubled by 2050 [5]. The correlations among temperature, humidity, poor air quality, health issues, and benefits have been proved in recent studies [6,7]. In a recent Spanish study, the effect of climate change was shown to increase the annual air pollution deaths and reduce the health benefits by 10% [8]. Moreover, a significant correlation was found between extreme heat stress and poor air quality conditions with increased concentrations of O₃, NO₂, and SO₂ [9]. Therefore, applying heat and air pollution mitigation strategies that protect human health and improve the urban environment is essential for urban dwellers [10]. Increasing the urban greenery, which includes vegetation, tree canopy, and building vegetation elements, influences the urban canyon thermal environment and the air quality as an evolutionary and efficient UHI mitigation [11,12,13]. Meanwhile, preserving and maintaining the greenery cover of the protected natural land, the so-called Greenbelt, and river valleys in Ontario helps in decreasing the ambient temperature and biogenic emissions and promotes the increase in evapotranspiration and shading.

Most of the reviewed studies ensured the great potential of increasing urban greenery cover for reducing ambient temperature and enhancing air quality levels. However, the holistic impacts of poor air quality and heat events on environmental and community responses are not well established. This paper aims to prove the association between ambient conditions and air quality variables within the GTA and reflects the effects of adaptation and mitigation strategies on environmental and community responses. The paper uniquely discusses the expected air quality and heat-related responses regarding the proposed enhancements in the urban microclimate. These responses holistically include the environmental, energy, health, and economic benefits. This study approaches the linkage among air quality, weather conditions, and health to holistically interpret the potential of mitigation strategies to improve the urban microclimate. A novel integrated statistical–simulation approach was developed to test the benefits of increasing the greenery cover around the GTA to the urban microclimate, energy, and community responses. Following a holistic perspective, the multiple impacts of air quality and ambient conditions within the GTA were discussed. The results can be used by decision-makers to initiate policies to improve living conditions for urban dwellers.

2. Materials and Methods

The method combined the statistical approach with microclimate simulations to investigate the impact of increasing urban greenery cover on the air quality and heat-related health responses. Firstly, the study intended to correlate the behavior of the meteorological parameters with the air quality variables to investigate the effect of heat-related variations on the concentration of air pollutants in the region of study. The proposed study parameters included meteorological parameters (air temperature and relative humidity) and air pollutant concentration (ground-level ozone, O₃, and fine particulate matter, PM_2.5). A statistical approach was introduced, including regression analyses for establishing the correlation between the study parameters. Then, microclimate simulations were designed using a developed and validated simulation code to assess the impacts of intensifying the greenery cover on the urban microclimate. The outputs of the predictive regression analysis and the microclimatic simulations were linked to concluding the impacts on air quality and human health response in the GTA.

Daily mean values of the meteorological and air quality variables were considered to assess the effect of meteorological parameters’ variations on air quality levels in the study region. The concentration of a single pollutant was considered as the dependent variable of the regression analysis. The independent variables of the model included the meteorological parameters and the concentration of other pollutants that can affect the correlation. The statistical model was built in both Microsoft Excel and JASP software to analyze the provided dataset using multiple regression analyses. The designed model provided evidence-based correlations between the parameters, promoting the predictive potential for further estimations. Before establishing regression modeling, critical regression assumptions diagnostics were conducted, including correlations between model variables, multicollinearity, autocorrelations, and variance. The model was adjusted to satisfy essential regression assumptions.

The modeling approach included investigating the cooling and warming effects associated with the increase in urban green cover. The study utilized an updated version of the microclimate simulation code of the Urban Weather Generator (UWG) that was initially created by Bueno et al. [14]. The UWG predicts the microclimate changes in a selected urban environment compared to the weather data from a nearby rural/airport weather station to assess the effects of the UHI on a local neighborhood scale. The multi-layered, three-dimensional code considers dynamic surface temperature, shortwave and longwave radiation fluxes, and sensible heat fluxes from roofs, walls, and roads. The model aggregates the fluxes into the exchange of momentum and energy between the urban surface and atmosphere. The model was improved by the authors to update the mitigation effect of the greenery cover considering the shading and evaporation effects on the urban microclimates. Additionally, new urban and building features were developed to allow the integration of novel mitigation techniques, such as green roofs and vegetated facades. All the details of the model development and validation can be found in [15]. Thus, the model could assess the mitigation scenarios associated with the urban green strategies which include tree canopy, vertical vegetation façades, and green roofs.

According to the Canadian Council of Ministers of the Environment (CCME), O₃ and PM_2.5 are the first two pollutants that are of concern to the air quality management system in Canada. Referring to the CCME’s air quality report, exposure to O₃ and PM_2.5 causes respiratory symptoms and reduced lung and heart function, with an increasingly high risk of emergency cases of respiratory or cardiovascular issues for sensitive populations (children, the elderly, and people suffering from chronic diseases) [16]. The study domain, shown in Figure 1, includes the Peel region, focusing on Mississauga (Mis) and Brampton (Brm) municipalities. Mississauga is characterized by higher urban densities, and Brampton is more characterized by the dispersal of river valleys.

The air quality and weather data were obtained from the Ministry of the Environment, Conservation, and Parks, Ontario [17] and Environment Canada [18], respectively, for 12 years from 2006 to 2017. This study only focused on warm and hot seasons, which extend from May to September each year. The historical hourly concentrations for O₃ and PM_2.5 were collected from Mississauga and Brampton air monitoring stations (43.55 N 79.66 W, and 43.70 N 79.78 W, respectively), shown in Figure 1. The weather data were obtained from the international airport weather station (43.67 N 79.63 W), which represents the only available historical weather data in the region of study that cover the intended period of study. Mississauga and Brampton air monitoring stations are located 14.4 km and 11 km, respectively, away from the weather station. However, all the stations share the same urban surroundings in terms of urbanity level and urban texture. The air monitoring stations provide hourly observations; thus, daily mean concentrations were calculated for the selected period. The daily weather parameters include mean values for ambient temperature and mean values for relative humidity.

3. Impact of Meteorological Variations on Air Quality

The regression analysis was conducted including the pollutant’s daily mean concentration as a dependent variable, while the mean temperature, relative humidity, and the other pollutant’s mean concentration were included as model variables. The results of the multiple regression analyses for both pollutants in Mississauga are presented in

Table 1. Regression analyses for pollutants’ average concentration in Mississauga.

Dependent Variable: O3_MIS_AVR				Dependent Variable: PM_MIS_AVR
Model	Coef.	t	p-Value	Model	Coef.	t	p-Value
(Intercept)	32.411	26.356	<0.001	(Intercept)	−12.320	−16.104	<0.001
RH_avr	−0.255	−17.294	<0.001	RH_avr	0.086	9.826	<0.001
PM_Mis_avr	0.736	19.004	<0.001	O3_Mis_avr	0.234	19.004	<0.001
Mean Temp	0.338	7.670	<0.001	Mean Temp	0.429	18.584	<0.001

. It can be inferred by the coefficients of the variables that the O₃ concentration is more correlated with the variations of PM_2.5 concentration, and with the mean temperature at the second place. With a significant p-value, the model shows a significant positive correlation between O₃ average concentration, mean temperature, and PM_2.5 mean concentration. Meanwhile, a significant negative correlation is observed between O₃ average concentration and RH mean value. Considering the PM_2.5 average concentration as a dependant variable, the model shows a significant positive correlation with O₃ average concentration, mean temperature, and RH mean value. It is noticed that the most effective variable in predicting the PM_2.5 concentration is the mean temperature. In both models, it is noticeable that the RH mean value is the least influencing variable on predicting the pollutants’ concentration. The predictive regression equations can be expressed by Equations (1) and (2) for predicting O₃ and PM_2.5 mean concentrations, respectively. It can be concluded that for every degree Celsius increase in the mean temperature, giving other variables as constants, the O₃ and PM_2.5 average concentrations are expected to increase by 0.34 ppb and 0.43 µg/m³ on average, respectively.

O₃ avr. conc. = 0.338 T_mean + 0.736 PM_{2.5 avr} − 0.255 RH_mean + 32.411,

(1)

PM_2.5 avr. conc. = 0.429 T_mean + 0.234 O_{3 avr} + 0.086 RH_mean − 12.32,

(2)

The above equations were used to predict the average concentrations of the pollutants in Mississauga. The predictive regression was conducted for two months (August and September 2016) covering two periods of heat warnings (4 August to 13 and 6 September to 8). Figure 2 and Figure 3 show the predictions of O₃ and PM_2.5 mean concentrations against the daily mean temperature. The inclination of the linear predicted regression lines confirms the significant correlation between the mean temperature and the pollutants’ mean concentrations. The figures show that the highest values for the O₃ and PM_2.5 mean concentrations were recorded during the August heatwave. Moreover, the peaks of the graph behavior for both pollutants followed the behavior of the mean temperature, which ensures the correlation between air quality and hot ambient conditions.

The regression analysis of the mean concentrations of both pollutants in Brampton is presented in

Table 2. Regression analyses for pollutants’ average concentration in Brampton.

Dependent Variable: O3_BRM_AVR				Dependent Variable: PM_BRM_AVR
Model	Coef.	t	p-Value	Model	Coef.	t	p-Value
(Intercept)	34.036	28.474	<0.001	(Intercept)	−13.298	−16.604	<0.001
RH_avr	−0.285	−19.842	<0.001	RH_avr	0.111	12.133	<0.001
PM_Mis_avr	0.724	19.911	<0.001	O3_Mis_avr	0.257	19.911	<0.001
Mean Temp	0.518	12.561	<0.001	Mean Temp	0.310	12.637	<0.001

. Comparing the results with those of the municipality of Mississauga, it is confirmed that the correlations and the effective variables are identical in both municipalities. As reported in Mississauga, the correlation is more significant between the mean temperature and the PM_2.5 mean concentrations. Overall, the predictive regression results are close to the predictive regression equations for Mississauga, which can be verified to predict pollutants’ mean concentrations in the GTA.

4. Effect of Increasing Urban Greenery Cover

The cooling effect of the urban greenery cover is due to the fraction of the blocked solar radiation that reaches the urban surfaces [11,12] and the evapotranspiration of plants and soil of the vegetation and tree coverage [19]. However, the increased humidity levels due to evaporation can affect thermal comfort, causing a counter warming effect. Most of the reviewed studies utilized numerical simulations to predict the effect of increasing the greenery cover [11,13,20]; however, the prediction of the warming effect and increased relative humidity due to the evaporation of the increased green cover is limitedly discussed. Additionally, a more accurate representation of the energy and mass balance for the urban air volume is required. This ensures the importance of developing a simulation tool that efficiently estimates the cooling effect of the green cover while considering the sequences of the warming effect on the urban climate.

The microclimate simulations were designed utilizing the developed and validated version of the UWG [15] to evaluate the current natural surroundings within the proposed neighborhood and to assess the effects of increasing the green and vegetation cover on the urban thermal behavior and building energy performance. Specifically, the study investigated the effects of preserving and developing the natural content of the river valleys region in Brampton, ON. A typical urban typology adjacent to a river valley was selected representing the Brampton residential neighborhood. The weather data for Guelph town, ON. was used as the rural weather station for simulation. The verification of the selection of the rural site, comparisons with nearby rural locations, and full characteristics of the rural site were provided by Dardir and Berardi [15]. The simulation was extended for 10 days (from 27 June to 6 July 2018) to include the reported heatwave (from 29 June to 5 July 2018) during this period. Details of location, urban features, and assumptions are provided in

Table 3. Specifications of the urban microclimate parameters.

Location	Brampton, ON
Distance from rural	35.5 km
Site area	528,000 m2
Building footprint	24.6% (130,000 m2)
Avr. building height	6 m
V-to-H1 ratio	0.25
Road urban area	33.5% (133,500 m2)
Water surface area	5.7% (22,500 m2)
Tree canopy	20%
Vegetation cover	60%
Building types	100% residential

¹ Vertical to horizontal aspect ratio for urban canyon.

.

The study mainly focused on air temperature and relative humidity as simulation outputs; it also monitored the energy consumption of the buildings. To assess the effects of increasing the greenery cover on the local urban climate of the Brampton neighborhood, three levels of the investigation were designed: increasing the tree canopy from an initial value of 20% of the urban area by 10% for three steps, integrating vertical vegetation façade systems that increase by 20% of the building façade area for three steps associated with largest tree coverage, and proposing green roofs coverage that increases by 20% of the building roof area for three steps associated with the maximum tree canopy and the vegetated façade system. As the urban area is limited by roads area and urban features, the tree canopy was selected to have a maximum possible coverage of 50% of the urban area. However, with an increased possibility to implement green roofs and vegetated façade systems, the maximum possible coverage was set to 60% of the building area. The stepped investigation was planned to evaluate the most effective mitigation strategy for reducing the ambient air temperature and building energy consumption. The green roofs involve integrated vegetated areas and plantation to building roofs which provide both insulation and shading; both effects were included in the developed version of the UWG. The results of the mitigation scenarios were compared with the current condition that was considered as a reference case. The reference conditions present 20% urban tree canopy, 0% green façade systems, and 0% green roof installations. All the results of the proposed scenarios for all strategies are presented in

Table 4. Ten-day average values for the simulation parameters.

		Air Temp (°C)	RH (%)	E (kWh/m2)
Reference case		22.19	71.88	6.05
Increasing tree canopy	30%	21.95	72.76	5.84
	40%	21.74	73.6	5.74
	50%	21.5	74.45	5.68
Increasing green façade system 1	20%	21.37	74.98	5.36
	40%	21.22	75.53	5.14
	60%	21.07	76.09	4.77
Increasing green roofs 2	20%	20.76	77.2	4.66
	40%	20.47	78.17	4.57
	60%	20.2	79.06	4.5

¹ Applying 50% of tree canopy. ² Applying 50% of the tree canopy, and 60% of green façade coverage.

.

Referring to the comparisons between the reference case and the scenario of maximum enhancements (50% of the tree canopy, 60% of green façade coverage, and 60% of green roofs), it was inferred that the 10-day average air temperature was reduced by 2 °C. Regarding the energy consumption of the buildings, applying the enhanced scenario saved an amount of 1.55 kWh/m² of energy consumption, which represents 25.6% of the energy consumption during 10 days of operation. Regarding the warming effect of the increased green cover, the increased relative humidity was tracked. The scenario with maximum enhancements achieved an increase of 7.2% of the 10-day average RH more than the reference case. To justify the effect of this increase on the thermal environment, the outdoor heat stress index was used to assess outdoor thermal resilience. It was expressed by the Canadian temperature–humidity index (Humidex). Humidex values were calculated based on the ambient air temperature (T_a) and dew point temperature (T_d) [21,22], as shown in Equation (3).

Humidex = T_a + 0.5555 × (6.11 × e^(5417.753 × (1/273.15 − 1/(273.15 + T_d))) − 10),

(3)

Humidex values were assessed for the whole simulation duration and are presented in Figure 4. The results show a reduction in the 10-day average Humidex from 27.6 to 25.7. During the heatwave period, the maximum Humidex value was reduced from 41.2 to 37.6. According to Health Canada [22], heat warnings are issued when the Humidex value exceeds 40, which was achieved during the heatwave applying the reference application. With the enhanced scenario application, the overall outdoor thermal performance was enhanced, and the condition of the heat warning was not met during the heatwave, keeping the neighborhood away from heat stress dangers. This ensures the effectiveness of the mitigation strategies in controlling extreme weather conditions, promoting climate resilience to the neighborhood population.

Regarding assessing the individual impact of the mitigation strategies, performance analysis was conducted for the proposed strategies on the air temperature and energy consumption for each step during the simulation. It is worth mentioning that while the value of each strategy was changed, the other strategies were constant. Figure 5a shows how far the mitigation strategies contribute to the evolution of canyon air temperature. As noticed, the tree canopy was the most effective strategy for reducing the canyon air temperature with a reduction of 0.23 °C for each 10% increase in the tree canopy, as derived from graph slope. Meanwhile, reductions of 0.08 °C and 0.14 °C were associated with each 10% increase in the green façade system and green roofs, respectively. Referring to Figure 5b, it can be concluded that the green façade systems were the biggest contributors to the building energy savings with an amount of 0.15 kWh/m² for each 10% increase in façade vegetation. Moreover, each increase of 10% in the tree canopy and green roofs contributed to the building energy savings by 0.08 kWh/m² and 0.04 kWh/m², respectively.

5. Health and Economic Responses

In this section, the linkage among ambient thermal conditions, air pollutants levels, population health response, and possible economic savings is approached. Based on statistics Canada [23] and Region of Peel [24], the annual all-cause mortality cases in Mississauga and Brampton are expected to reach 3530 and 3290 persons in 2030, respectively. Regarding the heat-related health response of the population, based on the correlations between all-cause mortalities and ambient temperature conducted by Anderson and Bell [25], an increase in mortality cases of 4.5% was estimated for each degree Celsius increase in ambient temperature. Using this estimation, the annual all-cause mortality cases are expected to increase by 159 and 148 persons in Mississauga and Brampton, respectively, for each degree Celsius increase in ambient temperature. Referring to the simulation results of the maximum enhancements of the urban microclimate, it can be inferred that, by increasing the urban greenery cover, the ambient air temperature was reduced by up to 2 °C. Thus, applying the rate that relates the mortality cases to ambient temperature, the reduction in all-cause mortalities in 2030 is expected to reach 311 and 290 persons in Mississauga and Brampton, respectively. This reduction in the total mortalities is estimated as 8.8% in both Mississauga and Brampton. Additionally, the reduction in ambient temperature is associated with improving the air quality levels, as indicated by the regression analyses. Based on the predictive regression and simulation results, the O₃ and PM_2.5 daily mean concentrations are expected to decrease by 0.68 ppb and 0.86 µg/m³, respectively, on average. These rates were considered to be applied to the municipality level, assuming the same enhancement procedure will be conducted for all the districts.

The economic benefits associated with reducing ambient air temperature were discussed by Wilson [26], applying the case study of Brampton. Avoiding exposure to extreme heat can save the total related losses by up to 45%. The related economic benefits include reducing the health system costs that involve the cost of hospitalization, increased emergency department visits, and ambulance calls due to heat exposure. They also include reducing the productivity losses attributed to increased breaks for outdoor workers due to heat exposure. Referring to the simulation results, the enhancements of the urban microclimate ensured that the neighborhood would not meet the conditions of heat warnings. Accordingly, applying the enhancement procedure for the GTA can maximize the economic benefits by higher potential savings of health and labor systems losses, solving many climate stress issues witin the GTA [4].

6. Conclusions

An integrated statistical–simulation approach was designed to figure out the influences of the adaptation of urban greenery cover on air quality and community responses. A newer version of the Urban Weather Generator (UWG) was used to assess the UHI mitigation strategies, including increasing the tree canopy (from 20% to 50%), integrating vertical vegetation façade systems (up to 60%), and incorporating moderate-to-intense green roofs (up to 60%). The results proved that the tree canopy and the green façade systems were the most effective strategies for reducing the canyon air temperature and building energy savings, respectively. Regarding building energy consumption, applying the enhanced scenario achieved energy savings of 25.6% of the energy consumption during 10 days of operation. Moreover, a statistical-based model was designed to find the correlations between the meteorological parameters and the air quality variables. The statistical approach included regression analyses for establishing correlations to assess the impact of meteorological variations on air pollutants concentrations. The results confirmed a significant positive correlation between O₃ and PM_2.5 concentrations and ambient temperatures. It was concluded that for every degree Celsius increase in the mean temperature, the O₃ and PM_2.5 mean concentrations were expected to increase by 0.34 ppb and 0.43 µg/m³ on average, respectively. Additionally, the paper approached the linkage to health and economic responses. Applying the heat-related health response within the region of study, reductions of 8.8% in all-cause mortalities were expected in the region of study in 2030. Additionally, potential savings of health and labor system losses were expected due to protecting the urban environment from extreme heat exposure. The integrated approach developed in this paper can be used by decision-makers to predict the appropriate urban greenery cover that maximizes the environmental, health, and economic benefits. Thus, specific policies can be initiated to improve living conditions for urban dwellers.