Canopy density effects on particulate matter attenuation coefficients in street canyons during summer in the Wuhan metropolitan area

Highlights

• Trees morphological features affected the dispersion of particulate matter in street canyons.

Abstract

Changes in vegetation traits influence the particulate pollution mitigating effects of trees in street canyons; however, it remains unclear whether tree canopy density (i.e. the proportion of the street floor covered by the vertical projection of the tree canopy) promotes or reduces this effect. A 12-day field experiment was conducted in four representative street canyons to examine the mitigating effects of street trees on particulate matter (PM) for PM₁, PM_2.5, PM₄, PM₇, PM₁₀, and total suspended particles (TSP) among four canopy density treatments, including (1) open spaces and areas with (2) sparse (≤35%), (3) medium (35–70%) and (4) dense (≥70%) canopy densities.

The results showed that canopy density is the dominant vegetation trait that affects PM dispersion, with peak decreases occurring at a canopy density of ~30%. The particulate matter attenuation coefficient (PMAC) indicates the PM reduction capability of trees. The PMAC of each particle size class correlated negatively with canopy density and TSP (<100 μm) showed the greatest attenuation. In relation to open space treatment, a canopy density range 30–36% showed the largest reductions in the PM₁₀ and TSP concentrations of 26.75% and 27.49%, respectively. And for the PM_2.5 concentration, a canopy density range 24–36% exhibited the largest reduction (7.44%). It was also concluded that sparse canopy density is optimal for trees in areas with high PM concentration. Medium canopy density also promotes pollutant dispersion (especially PM_2.5), while dense canopy density causes air quality deterioration. This study will provide new insights into the response of atmospheric PM spatial dispersion to the characteristics of tree cover in street canyons, as well as the regulation mechanism of this response. By investigating this issue under different scenarios, this study aims to contribute to the quantitative tree planting design in urban planning.

Introduction

Atmospheric particulate matter (PM), such as PM_2.5 and PM₁₀, poses a risk to human health (Bouhuys et al., 1978; Samet et al., 2000). Inhalation of PM can cause human health concerns, including lung cancer, asthma attacks, and cardiovascular diseases (Seaton et al., 1995; Lelieveld et al., 2015). In urban areas, relatively narrow street environments surrounded by continuous lines of tall buildings on both sides of the street can be called street canyons (Vardoulakis et al., 2003; Xue and Li, 2017). Motor vehicle traffic is one of the main sources of urban air pollution and air pollutant concentrations are typically high in street canyon environments with large traffic volumes (Johnson et al., 1973; Jeanjean et al., 2016). The low vertical source location of motor vehicle emissions causes air pollution exposure to pedestrians walking in street canyons or to residents of the lower floors in buildings facing the street canyon (Baldwin et al., 2015; Farrell et al., 2015; Rafael et al., 2018), causing adverse health effects to the exposed individuals (Lelieveld et al., 2015; Gao et al., 2017). It has been suggested that trees contribute to urban air purification through their ecological functions, such as air pollutant adsorption by leaves, tree canopy dust retention, and the release of oxygen (Chen et al., 2015; Gomez-Moreno et al., 2019). In urban planning and environmental management, effective use of urban trees to reduce vehicular traffic-related air pollutants remains an important challenge.

However, it is not clear in what circumstance trees in street canyons or near-road environments have positive or negative impact on air pollutant concentrations (Gromke et al., 2008; Vos et al., 2013; Tong et al., 2015; Buccolieri et al., 2018). Previous studies have suggested that the impact of tree cover on PM concentrations also depends on the particle size classes (Gromke et al., 2008; Yli-Pelkonen et al., 2017; Viippola et al., 2018). For instance, Viippola et al. (2018) found that evergreen forests near roads slightly increase the local PM_2.5 concentrations, but coarser particle pollution can be reduced by evergreen forest vegetation. Nevertheless, it is believed that street trees can mitigate PM pollution by adsorption and deposition due to the large canopy surface area of their leaves, stems and branches (Lovett, 1994; Beckett et al., 2000; Nowak et al., 2006), especially in near-road environments (Islam et al., 2012; Setälä et al., 2013; Baldauf et al., 2017; Yli-Pelkonen et al., 2017). It has been reported that trees have a regionally beneficial impact on road traffic emissions and reduce ambient concentrations by 7% at pedestrian height (Jeanjean et al., 2015). Furthermore, Yli-Pelkonen et al. (2017) showed that the coarse particulate levels in tree-covered areas were 23% (p = 0.023) lower than in treeless near-road areas, indicating that urban tree planting can reduce particulate pollution levels in certain size classes (see also Salmond et al., 2016). On the other hand, a number of recent studies have demonstrated that trees in urban street canyons generally result in increased traffic-emitted gaseous pollutants and PM concentrations (Gromke et al., 2008; Gromke and Ruck., 2008; Buccolieri et al., 2009; Amorim et al., 2013; Vos et al., 2013; Morakinyo and Lam., 2016b; Xue and Li, 2017). The reason for this phenomenon is the tree crown flow resistance, wherein the trees reduce the air exchange rate by reducing the wind velocity and increase turbulence by obstructing canyon eddy conditions (Gromke et al., 2008; Salim et al., 2011; Gromke and Ruck, 2012; Moonen et al., 2013). Consequently, the dispersion of near-ground emitted traffic pollutants is limited, causing these pollutants to accumulate within the canyon, particularly below the crown level of dense street tree stands (Vos et al., 2013; Gromke and Blocken, 2015). The highest pollution levels are created with vegetation under oblique winds, and the tree crown flow resistance effect is more pronounced in configurations with poor ventilation, such as low wind speed, perpendicular inflow direction, and deep canyons (Gromke and Ruck, 2007; Buccolieri et al., 2011; Wania et al., 2012). Concerning air quality, street trees resulted in increased neighborhood-averaged PM concentrations, showing a 1% increase relative to the treeless environment per percent crown volume fraction (Gromke and Blocken, 2015). Therefore, air pollutants accumulating below tree canopies may result in higher PM pollution levels in street canyons with trees (Gromke et al., 2008; Moonen et al., 2013; Abhijith and Gokhale, 2015).

These adverse effects of street canyons emphasize the need to control vegetation variables. Prior studies have noted the importance of vegetation on influencing the PM concentrations in street canyons (Gromke and Ruck, 2008; Xue and Li, 2017; Tiwari et al., 2019). With respect to the degree of influence, it was found that canopy density (Wania et al., 2012; Jin et al., 2014; Gromke and Blocken, 2015; Abhijith et al., 2017), row spacing (Ozdemir, 2019), hedge height and width (Al-Dabbous and Kumar, 2014), leaf area index (LAI) (Jin et al., 2014; Morakinyo and Lam, 2016a), species characteristics (Yang et al., 2015; Leonard et al., 2016) and planting configuration (Chen et al., 2016.; Morakinyo et al., 2016a; Abhijith and Kumar, 2019) had the greatest impact among the various vegetation traits. Research has been carried out on some of these traits; however, a study investigating which vegetation trait has the predominant impact on PM concentrations has not been performed.

Certain planting patterns can improve the local and near-road air quality (Baldauf, 2017; Gomez-Moreno et al., 2019; Tiwari et al., 2019). Some related studies suggest that custom-designed street trees can change the microclimate conditions, particularly increasing the airflow, which promotes airborne particle dispersion and improves the air quality in streets (Memon et al., 2010; Yassin and Ohba, 2012; Buccolieri et al., 2018). Among these studies, only a small number included numerical modelling focused on the aerodynamic effects of canopy density on the dispersion of different PM types (Buccolieri et al., 2009; Wania et al., 2012; Moonen et al., 2013). The wind speed under thick canopies is known to be lower than that in open, treeless areas (Gromke et al., 2008; Wania et al., 2012; Setälä et al., 2013; Buccolieri, 2018), which likely prevents road dust from being dispersed further away from roads in tree-covered areas (Yli-Pelkonen et al., 2017). However, current research on canopy density is still scarce, which highlights the need for a detailed investigation of the effects of trees crown sizes on air quality (Abhijith and Gokhale, 2015; Tiwari et al., 2019). It is essential to use quantitative indicators, such as canopy density and LAI to study the influence of tree planting on PM levels. As such, to enhance the understanding of pollutant dispersion due to urban vegetation, additional on-site, small-scale measuring experiments are needed (Yli-Pelkonen et al., 2017; Tiwari et al., 2019).

Planting trees in street canyons is controversial because it is still unclear how trees configuration affects PM dispersion in the street canyon environment. This study hypothesized that an optimal range of tree canopy density in street canyons exists to reduce PM concentrations. More specifically, the following research questions were investigated: 1) which vegetation traits have the largest impact on PM concentrations in street canyons? 2) is there a direct correlation between the particulate matter attenuation coefficient (PMAC) and canopy density (CD)? 3) which street tree planting configuration (sparse, medium or dense canopy density) results in the largest reduction of PM concentrations? To answer these questions, 12-h PM concentration monitoring experiments were performed in four representative street canyons over 12 days. Furthermore, data on building geometrics, vegetation variables, traffic volumes, and meteorological factors in the studied street canyons were obtained for the study.