Zone of influence for particle number concentrations at signalised traffic intersections
Graphical abstract
Introduction
Signalised traffic intersections (TIs) are considered as a hotspot of particle number concentration (PNCs). Over 99% of PNCs are represented by particles below 300 nm in diameter (Kumar et al., 2011, Kumar et al., 2010). Short-term exposure (i.e. exposure to peak PNCs averaged over short durations such as 1 s averaged PNC) at the TIs contribute disproportionately higher exposure compared with those experienced at the rest of a commuting route with free–flow traffic conditions. For instance, our recent work found that as little as about 2% of commuting time spent in car at TIs can contribute up to about 25% of commuting exposure (Goel and Kumar, 2015). Furthermore, epidemiological studies have shown that even a short exposure of healthy people to traffic–emitted nanoparticles can cause reduction in brain plasticity (Bos et al., 2011) and induce changes in biomarkers of pulmonary and systematic inflammation of healthy individuals (Jacobs et al., 2010).
To assess the effect of changes in driving condition of a vehicle due to traffic signal on PNCs, a number of studies are carried out at the fixed monitoring sites around the TIs (Holmes et al., 2005, Morawska et al., 2004, Tsang et al., 2008, Wang et al., 2008). For instance, Tsang et al. (2008) assessed the pedestrian exposure to PNCs at a busy TI in Mong Kok, Hong Kong. They observed a sharp increase in PNCs as a result of vehicle acceleration after about 3 s when the traffic signal colour changed from red to green. Wang et al. (2008) found that average PNCs at a TI during red–light period was nearly 5–times higher compared to those during green–light period. Aforementioned studies were conducted at a distance of 3–5 m away from the intersecting roads and PNCs has been found to decrease exponentially at distances perpendicular to the road (Al-Dabbous and Kumar, 2014, Fujitani et al., 2012). Therefore, these studies did not capture the actual on–road PNCs. Moreover, most of past studies are carried out at fixed sites at a point near a TI that made it challenging to capture the on–road profile of PNCs on intersecting roads and at longitudinal distances from the centre of TIs. A number of studies have also carried out mobile monitoring of nanoparticles within the vehicles (Goel and Kumar, 2015, Hudda et al., 2011, Joodatnia et al., 2013a, Joodatnia et al., 2013b, Knibbs et al., 2010, Zhu et al., 2007), but studies focussing on the on–road profiles of PNCs around TIs are still scarce and covered as a part of this study.
There is a certain longitudinal distance along the road at both sides from the centre of a TI that experiences elevated level of exhaust emissions due to interruptions in traffic flow produced at the traffic signals. We refer this affected longitudinal length of the road as a zone of influence (ZoI) of a TI. The pollutant concentration in this zone can be many times higher as compare to rest of the route. For instance, Kim et al. (2014) observed that ZoI of a four–way TI for oxides of nitrogen (NOx) extends from −200 to 200 m distance from the centre of a TI in stop– and go– driving conditions. They found about 200–1000 ppb of additional NOx was observed within the ZoI compared with the rest of the route length. This is currently unknown whether the similar increase in PNCs can be expected within a ZoI of a TI. Our recent work suggested up to 29–times higher PNCs at TIs than those found on the rest of the route during free–flow traffic conditions (Goel and Kumar, 2015). This indicates much higher PNCs at different types of TIs but what is the ZoI under different driving conditions is yet poorly understood – this is one of the aims of this study.
As highlighted in our recent review (Goel and Kumar, 2014), dispersion modelling of nanoparticles at TIs is challenging due to a complex interplay among emission, dispersion and transformation processes. A number of operational air quality models addressing the dispersion of gaseous pollutants and particulate matter (PM) at TI are currently available, as seen from the review of related modelling studies in Table 1. However, no such dispersion model exists for the PNCs which could be used at the TIs. As a first step in this direction, one of the aims of this study is to derive a set of equations that represent PNC profiles within ZoI of four different types of TIs under frequently occurring stop– and go– driving conditions.
The distinctive features that aim to fill the existing research gaps of this work are as follows. Firstly, as opposed to previous studies (Holder et al., 2014, Hudda et al., 2011, Knibbs et al., 2010, Zhu et al., 2007) that have analysed the effect of velocity variations, ventilation settings and traffic conditions on in-cabin PNCs at individual commuting routes, this study has assessed the effect of traffic driving conditions on on-road PNC profiles at urban traffic hotspots (i.e., TIs). Secondly, this is for the first time when the ZoI of four different types of TIs under varying driving conditions (i.e. stop and go, multiple stopping, and free–flow) are defined for the PNCs. Thirdly, a set of equations representing the PNC profiles at different types of TIs are derived. The coefficients of these equations are represented in the form of parameters such as delay time, particle number flux, wind speed and driving speed. These parameters are chosen from the set of parameters on the basis of dimensional analysis. To perform dimensional analysis, a combination having dimensions similar to the dimensions of the coefficient of generalised PNC profiles is formed. Afterwards R2 value was estimated between the derived value of coefficient and the values predicted by using the above-noted parameters. If R2 value was less than 0.5, we discarded the proposed combination and tried other combinations. This process is explained in detail in Supplementary Information (SI) Section S1. Such equations could be useful for developing dispersion models for nanoparticles at TIs. Fourthly, this is the first time when positive matrix factorisation (PMF) has been applied to quantify the contribution of PNCs towards the total PNCs released during deceleration, acceleration, cruising and creep-idling at different types of TIs, as defined in Table 2. Such a quantification is useful to plan mitigation strategies to limit emissions and exposure to PNCs at pollution hotspots.
The overall aim of this study is to estimate ZoI and derive a set of equations describing PNC profiles within the ZoI at different types of TIs. Receptor modelling tool, PMF, has been applied to estimate the contribution of different driving conditions towards the total PNCs measured at the TIs and the rest of the route.
Section snippets
Route characteristics
Measurements were conducted on a 6 km long round route, which included 10 signalised TIs in Guildford, UK (Fig. 1). The route was chosen with an intention to pass through a maximum number of TIs of different geometries and built–up area around them so that varying impact of PNC dispersion at these TIs can be assessed (Goel and Kumar, 2014, Goel and Kumar, 2015, Patton et al., 2014). The pavement of each of the studied 10 TIs was made of bituminous flexible pavement, which contained ∼95% of
Results and discussions
Firstly we quantified the longitudinal distance of ZoI for PNCs at four different types of TIs (i.e. TI4w-nb, TI4w-wb, TI3w-nb and TI3w-wb) under three unique driving conditions. These included stop and go, multiple stopping, and free–flow driving conditions (Section 1 Introduction, 2 Methodology, 3 Results and discussions). Thereafter, PNC profiles within the ZoI of different types of TIs are interpreted based on dimensional analysis (Section 3.4). PMF analysis is then carried out to quantify
Summary, conclusions and future work
This study presents the longitudinal distances, representing ZoI around a TI, at four different types of TIs under three different driving conditions. ZoI of a TI was estimated on the basis of on-road PNCs and driving speed with respect to distance from the centre of a TI at different TIs. These on-road PNCs were obtained by mobile monitoring of particle number and size distribution in 5–560 nm size ranges. The results are discussed in terms of ZoI of a TI for PNC. Generalised profiles of the
Acknowledgements
Authors thank the UK Commonwealth Commission, Prof Alan Robins, Mr Shobhan Navaratnarajah, Mr Mihai Pop, Mr Ganesh Chandrashekran, Mr Santosh Tirunagari and the Department of Civil & Environmental Engineering at the University of Surrey, UK, for their help during experiments and discussion. PK thanks the funding received from the University Global Partnership Network (UGPN) through the project “Comparison of Air Pollution in Transportation ENvironments (CAPTEN): Development and Demonstration
References (51)
- et al.
The influence of roadside vegetation barriers on airborne nanoparticles and pedestrians exposure under varying wind conditions
Atmos. Environ.
(2014) - et al.
No exercise-induced increase in serum BDNF after cycling near a major traffic road
Neurosci. Lett.
(2011) - et al.
Ground-fixed and on-board measurements of nanoparticles in the wake of a moving vehicle
Atmos. Environ.
(2011) - et al.
A semi-empirical model for predicting the effect of changes in traffic flow patterns on carbon monoxide concentrations
Atmos. Environ.
(2003) - et al.
Measurements and predictors of on-road ultrafine particle concentrations and associated pollutants in Los Angeles
Atmos. Environ.
(2008) - et al.
Seasonal differences of the atmospheric particle size distribution in a metropolitan area in Japan
Sci. Total Environ.
(2012) - et al.
A review of fundamental drivers governing the emissions, dispersion and exposure to vehicle-emitted nanoparticles at signalised traffic intersections
Atmos. Environ.
(2014) - et al.
Characterisation of nanoparticle emissions and exposure at traffic intersections through fast–response mobile and sequential measurements
Atmos. Environ.
(2015) - et al.
A hybrid model for predicting carbon monoxide from vehicular exhausts in urban environments
Atmos. Environ.
(2005) - et al.
Performance evaluation of air quality models for predicting PM10 and PM2.5 concentrations at urban traffic intersection during winter period
Sci. Total Environ.
(2008)