Science of the Total Environment 505 (2015) 606–614

Contents lists available at ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

Traffic air quality index Zbigniew Bagieński ⁎ Institute of Environmental Engineering, Poznan University of Technology, ul. Piotrowo 3a, 60-965 Poznan, Poland

H I G H L I G H T S • The traffic air quality index (TAQI) was proposed as a tool of assessing air quality near roadways, especially in compact settlement areas. • Degrees of harmfulness were determined for primary pollution from traffic sources, which enabled describing the TAQI with a single value. • An example was provided of how the TAQI could be used to assess air quality in several real traffic routes in an urban area.

a r t i c l e

i n f o

Article history: Received 9 July 2014 Received in revised form 12 September 2014 Accepted 12 October 2014 Available online xxxx Editor: P. Kassomenos Keywords: Air quality Vehicle emission Degrees of harmfulness of pollutants Urban heat island

a b s t r a c t Vehicle emissions are responsible for a considerable share of urban air pollution concentrations. The traffic air quality index (TAQI) is proposed as a useful tool for evaluating air quality near roadways. The TAQI associates air quality with the equivalent emission from traffic sources and with street structure (roadway structure) as anthropogenic factors. The paper presents a method of determining the TAQI and defines the degrees of harmfulness of emitted pollution. It proposes a classification specifying a potential threat to human health based on the TAQI value and shows an example of calculating the TAQI value for real urban streets. It also considers the role that car traffic plays in creating a local UHI. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Air quality in an urban area is conditioned by complex interactions among numerous factors including climate, meteorology, physiographic, urban and social conditions and mainly air pollution emissions from anthropogenic sources. The most important property of air quality is its chemical composition and exposure level, particularly a share of substances hazardous to humans and the environment. Therefore, the term “air quality” is often limited to concentration levels of particular pollutants (e.g. Sokhi, 2008; Directive of the European Parliament, 2008; Regulation of the Minister of Environment (Poland), 2012). Over 70% of European cities do not meet air quality criteria recommended by the World Health Organization (WHO, 2006). As noted by Crutzen (2004), the rapid development of urban areas observed in the last several decades confirms that they have a considerable impact on air quality not only within a particular city, but also on continents and in the entire world. The urban ecological footprint is much bigger than the area taken up by an agglomeration. Urban areas that do not have ⁎ Fax: +48 616652439. E-mail address: [email protected].

http://dx.doi.org/10.1016/j.scitotenv.2014.10.041 0048-9697/© 2014 Elsevier B.V. All rights reserved.

many high-emitting industrial plants are primarily polluted by fuel combustion processes in stationary and mobile installations, which emit about 500 chemical compounds (Fenger, 1999). The author determined a mechanism of numerous primary and secondary air pollution substances affecting human health and the environment. The following were isolated: − Major Air Pollutants — MAP: sulfur dioxide (SO2), carbon monoxide (CO), lead (Pb), nitrogen oxides (NOx), particles PM (PM10, PM2.5, PM1.0), and ozone (O3), − Hazardous Air Pollutants — HAP: complex chemical, physical and biological substances of various types including mono- and poly-cyclic aromatic hydrocarbons. MAP are most often accepted as measurements of air quality in an urban area considering their impact on human health both in the general population and risk groups. Air quality standards are determined by boundary values of concentrations of these substances with regard to their impact on human health and natural and anthropogenic environments. Urban air quality assessment is a complex issue. The law in the European Union (Directive, 2008) and certain other countries (e.g.

Z. Bagieński / Science of the Total Environment 505 (2015) 606–614

Nomenclature CAQ CCOb CCO

category of air quality measured concentration of CO (μg m−3) concentration of CO after deducting the background (μg m−3) linear density flux of CO emitted from traffic sources ECO (μg m−1 s−1) density flux of an i pollutant emitted from traffic Eti sources per linear distance of roadway (mg km−1 s−1) or (kg km−1 year−1) density flux of an equivalent pollutant emitted from EtR traffic sources per linear distance of roadway (mg km−1 s−1) or (kg km−1 year−1) street canyon factor (m/m) — ratio of the building averFC age height to the total width of the street, HBav/W values of vehicle fuel consumption (kg s−1) FCv building average height (m) HBav degree of harmfulness of emitted pollutant i-th in relaki tion to SO2 L length of the street (m) N traffic volume — vehicle per hour (v/h) direct energy — total energy used by the vehicle in fuel Qd (kJ) flux of heat emitted from traffic sources per linear disQdL tance of roadway (kW km−1) heat of fuel combustion (kJ kg−1) QF TAQI traffic air quality index (mg km−1 s− 1) or (kg km−1 year−1) ux, uy, uz, components of wind velocity W width of the street (m) w vehicle velocity (km/h) coefficient considering the street canyon factor FC. YWC

EPA, 2003; Environment Canada, 2013) specifies the acceptable concentrations of some air pollutants, especially in cities and requires that concentrations be determined by measurement according to a unified method. Measurement results are the basis for preparing and publishing the Air Quality Index (AQI), which is mainly used to assess health risk among the inhabitants of an urban area (or a region). There are differences among the countries as to the methodology of determining the AQI and the considered pollutants (e.g. Air Quality Index EPA, 2003; Environment Canada, 2013; Guide to UK Air Pollution Information Resources, 2013). Measurements also enable to prepare forecasts and verify mathematical models. However, air quality assessment based on measurements in urban areas with a diversified building development and diversified low emissions is error-burdened when results are obtained at a background air pollution measurement plant. It also has limited representativeness when measurements from a local station are used. Air quality assessment according to mathematical deterministic systems has considerable uncertainty in characterizing meteorological, topographical and urban input data. Analyzing the relation between concentration values obtained after measurements as part of the Urban 2000 experiment and research in Los Angeles and Salt Lake City, Hanna et al. (2003) stated that the basic Gaussian dispersion model adapted to local meteorological and topographical conditions could explain a large part of variables in tests. However, concentration values in specific receptors even differ by a factor of 2 or 3. In compact settlement areas with stationary and mobile low emission sources pollution dispersion is stochastic, in a situation in which a small change in initial conditions causes a radical shift in concentration fields. There are many factors that can affect concentrations, for example, structure of

607

buildings, structure of streets, and topography of city area. Berkowicz et al. (2006) present an application of the Danish Operational Street Pollution Model (OSPM) to model values of traffic pollution concentration in cities based on emission values. A relatively high accordance of calculated and measured concentration values, especially for smaller emissions. Bagieński (2011) presented a new approach to air quality evaluation in an urban area, especially in compact settlement areas. It consists of basing assessment on non-random magnitudes (i.e. burdened with low uncertainty) and associating a result with its anthropogenic cause (i.e. dependent on humans). Identifying the cause makes it possible to correct it during a specified period of time, which means it is a dynamic element of air quality control. As stated earlier, in cities without many high-emitting industrial plants and with a generally low level of pollution coming from outside an urban area, the basic anthropogenic element that has a dynamic impact on air quality is fuel combustion in stationary and traffic sources. An important (anthropogenic) factor that influences air quality statically is the urban structure. Therefore, an air quality assessment index was proposed that associated air quality with emission size and structure and technical emission conditions ensuing from the development structure. Bagieński (2011) suggested a method of determining the Energy Air Quality Index (EAQIs) with regard to stationary combustion sources, which is particularly important in cities found in areas with cold and temperate climates during the so called “heating season”. This paper presents the methodology of determining the traffic air quality index (TAQI) with an example of its application. Many authors e.g. Baldauf et al. (2013) and Carpentieri et al. (2012) pay attention to the fact that vehicle emissions are responsible for a considerable share of urban air pollution concentrations. Numerous papers (e.g. Fenger, 1999; Park et al., 2004; Berkowicz et al., 2006; Clements et al., 2009; Kim and Guldmann, 2011) present the results of traffic pollution concentration measurements near varying traffic volume streets and analyze the range of the route impact. Research shows that concentration values are highly sensitive to dynamic anthropogenic conditions (traffic volume, vehicle type), urban structure and meteorological conditions. Air pollution dispersion in urban streets is a complex issue and depends on many factors, including: − Structure of buildings near streets — the height and the relative height of buildings, their shape, including the shape of their roofs. Buildings are responsible for the disturbances in the airflow. The boundary of the wake region (with horseshoe vortex) can approach up to 3 heights of a building and the range of 16 heights at the lee side of the building (Ahmad et al., 2005; Peterka et al., 1985). The cumulative effects of groups of buildings (of compact settlement) make local air flows differ considerably from a flow in an undisturbed area as to direction and velocity. − Structure of streets — their width, intersections and orientation in relation to the mostly frequent wind directions, but also with regard to, say, a group of trees in a particular street. − Geometry of street canyons — regular or irregular canyons, heightto-width ratio and length-to-height ratio of a canyon; the so called regular canyon only appears in models. Buildings surrounding real street canyons are diversified as to their height and shape and the type of their facade. − Vehicle motion — traffic intensity, vehicle velocity, one- or two-way traffic, and heat stream emission. The vehicle wake and the hot exhaust gases generate mechanical and thermal turbulence. This can have an influence on the airflow in the canyon up to 12 m above the road (Qin and Kot, 1993; Kaster-Klein et al., 2000). − Wind direction and velocity — many authors e.g. Chang and Meroney (2003), Ahmad et al. (2005), Chana et al. (2008), Kim and Guldmann (2011), and Carpentieri et al. (2012), pay attention to the considerable impact that wind direction and velocity have on flow distributions and pollutant concentration in a street canyon.

608

Z. Bagieński / Science of the Total Environment 505 (2015) 606–614

Table 1 Category of air quality dependent on the TAQI value.

Table 2 Degrees of harmfulness of emitted pollutants ki.

TAQI (mg km−1 s−1)

Category of air quality CAQ

Pollutant

SO2

NOx

CO

PM2,5

C6H6

NMVOC

CO2

0–1000 1001–5000 5001–12,000 12,001–20,000 Above 20,000

1 — Very good 2 — Good 3 — Moderate 4 — Unhealthy 5 — Hazardous

ki

1

5.6

0.13

9.1

44

4.4

1.5 ∙ 10−5

This influence depends considerably on the geometry of a street canyon and it is usually tested in wind tunnels. It can be generally assumed that as for the parallel wind the highest concentrations occur on the road and lower parts of the canyon. However, as far as the perpendicular wind is concerned, the highest vertical average concentrations occur near building walls (on the leeward side) (Park et al., 2004). In real conditions wind velocity has three, and not two, components ux, uy, uz, which change in time and which sum up with disturbances in the airflow generated by buildings (the surrounding of a canyon), street intersections, vehicle motion and other topographic elements and technical factors. Rodriguez et al. (2013) say as in the areas of compact settlement considerable fluctuations of wind direction and its velocity lead to a general velocity reduction. − Air convection in the wake of the local heat sources (local heat island) — apart from considerable heat emissions from vehicles, it can also be heat emitted from heated buildings, air conditioners and wall surfaces that absorb sun rays. It is important to assess a potential health threat to people living in high-volume traffic streets and answer the question how air quality can be improved, e.g. in many old European cities. Understanding dispersion in urban streets is a very important issue in urban planning and air quality management. These phenomena are often neglected in urban modeling studies. 2. Method of calculating the traffic air quality index — TAQI 2.1. Defining the traffic air quality index The traffic air quality index is defined by the product linear density flux of an equivalent pollutant emitted from traffic sources EtR and relative increase in the TAQI related to the H/W ratio of street canyon YWC. TAQI ¼ EtR  Y WC :

ð1Þ

Legend: −NOx = NO + NO2; over 90% of nitrogen compounds are emitted in the form of nitric oxide NO. −PM2.5 — in the total emission of PM10 from traffic (emission from combustion and friction), PM2.5 amounts to 80–85%, in which particles emitted with exhaust fumes equal 75– 80%, out of which elemental carbon (EC) is responsible for about 60%. −Benzene (C6H6) — amounts to about 5% of NMVOC stream; stream NMVOC is composed of MAHs in about 30%. −NMVOC — result from combustion and fuel evaporation from a hot engine, carburettor and fuel tank.

air quality and based on the TAQI values (Table 1) considering the results of analyzing streets in an urban area. The aim of TAQI is to enable assessing a potential threat to the health of people living in high-volume traffic streets, especially in compact settlement areas. It can be used to assess in what situations it is necessary to improve air quality in an area. The TAQI associates evaluating air quality near roadways with factors dependent on people's actions and decisions, namely emission from traffic sources and street structure. It is possible to calculate emission of particular pollutants based on characteristics of vehicles and their motion. The coefficient YWC describing street structure must include a lot of factors (mentioned in Introduction) which influence pollution dispersion in a canyon. This requires the following simplified assumptions be accepted: − A street canyon is regularly specified by the ratio of the average height of buildings to the total width of the street − Wind direction is perpendicular to a street − Disturbances in the airflow generated by buildings, street intersections, vehicle motion and other topographic elements and technical factors make it possible to average vertical concentrations on the surfaces of buildings delimiting the street as a selected range of the height of the street canyon concerns. Such assumptions may lead to an overestimation of health risks among inhabitants. It must be borne in mind, however, that phenomena such as vehicle pollution from other streets and periods of traffic congestion may cause threats greater than those described in the methodology, especially in urban compact settlement areas with poor ventilation. 2.2. Flux of pollutant emitted from traffic

In accordance with the principle accepted after the AQI was defined, high values of TAQI describe deteriorated air quality. The author proposes the classification of a category determining a potential threat to

YWC

Total emission from road transport is the sum of exhaust emissions, evaporative emissions and road vehicle tire and brake wear emissions. Exhaust emissions include hot and cold-start emissions. ETotal ¼ EHot þ ECold þ EEv :

3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

FC = HBav/W Fig. 1. Relation between coefficient YWC and street canyon factor FC.

ð2Þ

Hot emissions depend on the kind and amount of fuel burned and the kind and age of an engine, i.e. motion velocity and a vehicle's technical data such as the category of a vehicle (passenger cars, light- and heavy-duty vehicles, buses), engine capacity, the kind of fuel, and time of certification (permissible emission according to EURO standards). Cold-start emissions concern increased emissions of carbon monoxide and hydrocarbons and PM during the initial stage of the (cold) motion of a vehicle and depend on air temperature. In calculations they are treated as additional emissions considered solely for urban traffic. Evaporation emissions stem from fuel evaporating from a hot engine, carburettor and fuel tank. They are considered for fuel engine cars and depend on air temperature, humidity and fuel tank pressure related to these values. Vehicle emissions include pollutants such as CO, NO, NO2,

Z. Bagieński / Science of the Total Environment 505 (2015) 606–614

609

Table 3 Percentage share of vehicle type in car traffic scenarios. Variations

Passenger cars %

Trucks %

Petrol (gasoline) (l)

WIa WIb WIIa WIIb WIIIa WIIIb WIV

Diesel (l)

LPG

b1.4

1.4–2.0

N2.0

b2.0

N2.0

36 34 33 31 34 33 27

20 21 24 22 19 18 16

2 4 6 7 2 2 7

15 13 17 17 15 14 14

3 6 8 7 3 3 8

PM2.5, PM10, CO2, VOCs including mono- and poly-cyclic aromatic hydrocarbons (MAH, PAH) and benzene C6H6, and to CH4 and N2O, metals like Cd, Cr, Cu, and Ni. In the recent years particular attention has been paid to PM emission. Fine particles PM2.5 (diameter ≤ 2.5 μm) and ultra fine particle (diameter ≤ 0.1 μm) emissions in a lot of urban regions are mainly emitted by diesel vehicles (Kim and Guldmann, 2011). Directly emitted elemental carbon soot (black carbon) from diesel vehicles, re-suspended dust from roads, and secondary acidic particles are the major contributors of PM in urban roadside areas. The problem is particularly serious in cities with a large proportion of diesel-powered vehicles, e.g. Barcelona — about 50% (Schembari et al., 2013). According to the Air Quality in Europe — report (2013), between 2009 and 2011, up to 96% of city dwellers were exposed to fine particulate matter (PM2.5) concentrations above WHO guidelines. Emission value of i-th substance Eti for a street is determined as a sum of emissions for particular categories of vehicles (the kind of fuel, engine capacity, time of certification). Traffic volume for street considering types of vehicles for a specific period of time (considering hour-tohour and day-to-day variability) makes it possible to estimate based on measurement or e.g. using the SATURN software. Values of emission indexes for particular categories of vehicles and the methodology of calculating emission are specified in the EMEP/EEA air pollutant emission inventory guidebook (European Environment Agency, 2010) and TREMOVE — Final Report European Commission (2007). The guidelines are used in e.g. the COPERT IV computer program to calculate traffic emissions for various cases including roads both in an urbanized area and outside it. Depending on the scope of calculations, emissions of various pollutants can be analyzed, especially with regard to CO, NO, NO 2, PM2.5, PM10 , NMVOC, C 6H6 , and CO2 . PM emissions from road vehicle tire and brake wear are also taken into consideration. Emission values may be referred to the linear dimension Eti (mg km−1 s−1). The vehicle emission values can also be referred to a particular area of the city as the surface density flux of pollutant emitted from traffic sources Etia (mg m−2 s−1). Based on the defined degrees of harmfulness ki of analyzed pollution the flux of an equivalent pollution

13 12 12 15 12 11 8

Light-duty trucks

Heavy-duty trucks

Gasoline

Diesel

Gasoline

Diesel

5 6 0 0 8 8 0

5 4 0 0 7 7 0

0 0 0 0 0 0 1

0 0 0 0 0 5 19

emission for a roadway (street) is determined: EtR ¼ Σðki  Eti Þ:

ð3Þ

Due to great variances of emission pollutants in street canyons, the values of EtR are determined as average values in specified periods of time (e.g. 7 A.M. to 7 P.M. in a working day) and for particular street segments. Combustion vehicles in motion also emit a considerable heat stream. They have a significant share in the anthropogenic component of the Urban Heat Island. Heat emission is equal to a direct energy stream generated during fuel combustion: Q d ¼ Σð FC v  Q F Þ:

ð4Þ

The determinant of the UHI generated by vehicles QdL (kW km−1) with regard to a street is calculated using: Q dL ¼ Q d =L:

ð5Þ

Crutzen (2004) sees the Human Energy Production as one of the factors that influence air quality. Heat emission from anthropogenic sources (buildings, vehicles, industrial plants) contributes to generating UHI and leads to an increase in the local air temperature. Especially in deep canyons in the summer a rise in air temperature combined with increased air pollution concentrations has a considerable negative impact on air quality and health of inhabitants. Value QdL is especially crucial in summer, when combining natural and anthropogenic factors shaping the UHI has a considerable impact on air quality. Therefore, an additional factor is used for summer which considers limited airflow conditions in a street canyon QdL · YWC. 2.3. Factor of a street canyon considering emission conditions Urban systems are complex, which means it is impossible to fully consider their influence on pollution dispersion, especially in compact settlement areas. The purpose of introducing the street canyon factor

Table 4 Fluxes of substance emission, equivalent emission and TAQI for summer. Street

CO

NMVOC

NOx

PM2.5

Benzene

mg km−1 s−1 WIa WIb WIIa WIIb WIIIa WIIIb WIV

2876.1 4295.7 199.9 479.8 2581.3 3775.2 3242.7

CO2

YWC

g km−1 s−1 382.8 571.8 27.2 65.2 339.7 543.3 448.9

296.7 443.2 38.8 93.2 350.3 783.9 2730.3

16.7 25.0 1.7 4.2 17.6 37.7 83.7

18.0 26.9 1.3 3.2 15.4 22.9 16.9

108.0 161.5 11.4 27.3 104.6 178.5 364.9

EtR

TAQI

CAQ

13,997 14,633 743 1573 5559 23,290 19,199

4 4 2 2 3 5 4

mg km−1 s−1 3.0 2.1 1.7 1.5 1.2 2.7 1.0

4666 6968 437 1049 4632 8626 19,199

–

610

Z. Bagieński / Science of the Total Environment 505 (2015) 606–614

Table 5 Fluxes of substance emission, equivalent emission and TAQI for winter. Street

CO

NMVOC

PM2.5

NOx

Benzene

mg km−1 s−1 WIa WIb WIIa WIIb WIIIa WIIIb WIV

5212.2 7785.7 293.4 704.2 5322.3 7937.3 4138.2

560.2 837.0 32.2 77.2 511.8 800.0 477.5

320.0 477.9 40.6 97.5 376.3 822.9 2757.1

25.7 38.5 2.3 5.6 26.5 51.5 90.8

29.1 43.5 1.8 4.3 26.5 39.6 21.2

FC into the TAQI formula is to consider a relative deterioration of the dispersion conditions of emitted pollution as an outcome of unfavorable technical conditions of emission. F C ¼ HBav =W:

ð6Þ

The street canyon factor FC is the ratio of the average height of buildings delimiting the canyon to the total width W of the street. Depending on the value of the ratio FC, three types of airflow perpendicular to the canyon symmetry axis can be classified (Ahmad et al., 2005; Oke, 1988): − skimming flow for FC N 0.8, which is characterized by a significant reduction of air exchange inside the canyon, where FC N 0.8, − wake interference flow for FC = 0.20–0.8, where the stream is circulated inside the canyon and part of pollution is cleansed and transported to adjacent canyons (mainly within the wake region), − isolated roughness flow for FC b 0.20, when the influence of reciprocal interaction of the canyon walls on the interference can be ignored. The value of FC has a decisive influence on diffusion conditions of pollution from traffic sources, and also from adjacent low stationary sources (Bagieński, 2006; Chang and Meroney, 2003). The factor does not consider the influence that the kind of vehicle motion (velocity, number and directions of streams, various sizes) has on traffic pollution dispersion in a street canyon (Ahmad et al., 2005). Based on investigations of Chang and Meroney (2003) and Sini et al. (1996) an equation determining the dependence of the TAQI value on FC for wake interference flow (Eq. (7)) was developed and boundary conditions were specified for the other types of flow. a) for FC N 0.8, YWC = 3.0 — skimming flow b) for 0.20 b FC b 0.8 — wake interference flow Y WC ¼

CO2

YWC

g km−1 s−1

C ¼ 0:72 exp ð1:75 F C Þ: C WCb0:2

ð7Þ

EtR

TAQI

CAQ

19,356 20,238 863 1828 7750 30,709 19,846

4 5 1 1 3 5 4

mg km−1 s−1

119.4 178.5 12.3 29.5 115.7 195.1 376.8

3.0 2.1 1.7 1.5 1.2 2.7 1.0

6452 9637 508 1219 6458 11,374 19,846

c) for FC b 0.2, YWC = 1.0 — isolated roughness flow — the TAQI does not depend on FC. Fig. 1 shows the relation between coefficient YWC and the street canyon factor FC. Based on Eq. (7) an average value of relative pollutant concentration may be estimated in the region of the leeward wall horizontally restricted by the plane z = 0.4 HBav. It is a space inhabited by people. Investigations by Kim et al. (2012) also confirm the highest concentrations of traffic pollutants in this area. The reference concentration is the value of pollutant concentration in the area under investigation for HBav = 0.2 W, i.e. when the influence of a canyon can be neglected. 2.4. Degrees of harmfulness of emitted pollutants in evaluating air quality Degrees of harmfulness of pollutants ki with regard to the reference substance are determined for primary pollution considering the following factors: − the direct influence of the pollutant on humans, − the range of pollutant interaction also in various environment components, − the life span of an emitted pollutant, − the physicochemical processes involving the pollutants, − the impact of secondary substances arising from pollution. The degrees of harmfulness thus differ from the toxicity coefficients which are determined based on permissible values of substances in atmospheric air. The degrees of harmfulness were determined for the following significant pollutants emitted from traffic sources: NOx, CO, PM2.5, C6H6, and NMVOC. Values defined in Life Cycle Assessment (LCA) procedures were accepted as the basis for determining the degrees of harmfulness. They concern primary pollutants and consider the listed factors. Based on the LCA procedure, it was accepted that the potential of harmfulness (equivalent pollution emission) is equal to the sum of the products of substance emissions and its degree of harmfulness — vide Eq. (3). The

EtR (mg km-1s-1)

EtR (mg km-1s-1)

20000

20000 15000 15000 10000

10000

5000

5000 0

0 WIa

WIb

WIIa

WIIb

WIIIa

WIIIb

Fig. 2. Equivalent emission fluxes EtR — summer.

WIV

WIa

WIb

WIIa

WIIb

WIIIa

WIIIb

Fig. 3. Equivalent emission fluxes EtR — winter.

WIV

Z. Bagieński / Science of the Total Environment 505 (2015) 606–614

TAQI (mg km-1s-1)

611

Table 6 Values of determinants of UHI generated by vehicles.

35000

Street

30000

YWC

QdL kW km−1

25000

QdL · YWC

QdL

kW km−1

kW km−1

Summer

20000

WIa WIb WIIa WIIb WIIIa WIIIb WIV

15000 10000 5000 0 WIa

WIb

WIIa

WIIb

WIIIa

WIIIb

Winter

1594 2383 168 403 1545 2631 5367

3.0 2.1 1.7 1.5 1.2 2.7 1.0

4783 5005 286 605 1853 7105 5367

1762 2634 182 436 1708 2877 5543

WIV

Fig. 4. TAQI values — summer.

following environmental impact categories were selected (Guinee, 2001, Handbook on Life Cycle Assessment): − Human Toxicity Potential − HTP, − Photochemical Ozone Creation Potential — POCP.

of the discussed categories and standards. Some ki values may be subject to discussion based on other sources. 3. Example of determining the TAQI values for selected streets 3.1. Assumptions for calculations

The following ambient air quality standards were also analyzed: − permissible concentration of pollutants in the EC with regard to human health (Air quality in Europe, 2013), − air quality standards according to NAAQS — US EPA (Air Quality Index, 2003), − air quality standards according to the WHO (WHO, Air quality…, 2006), − air quality standards according to NAAQOs (Environment Canada, 2013), − permissible concentration of air pollutants in Poland according to the 13 September 2012 Regulation of the Minister of Environment (Poland). As usual, sulfur dioxide SO2 is assumed to be the reference substance. It is a pollutant whose emission is in fact independent of combustion conditions and is subject to slow change in atmospheric air. Other references in the LCA procedure are also used (3,4 dichlorobenzene for HTP and ethylene for POCP), which was taken into consideration. The value of ki was also calculated for CO2 as a pollutant which because of the magnitude of emission from combustion processes influences air quality in restricted diffusion areas. The degree of harmfulness for CO2 with respect to SO2 was determined based on their concentration in the air of medium-sized cities according to the European Standard for Ventilation EN 13779. Table 2 shows the values of degrees of harmfulness for main pollutants emitted from traffic sources. The value of degree of harmfulness results mainly from analysis

Assumptions were accepted based on real conditions of streets in Poznań, a Polish city with over 500,000 inhabitants. Street structure and car traffic variations: − WI — compact urban settlement, passenger traffic with a 10% share of light-duty trucks − WIa — a one-way traffic (two-line) street with the traffic volume of 1600 vehicles per hour (v/h), values of indexes: FC N 0.8; YWC = 3.0 − WIb — a two-way street with the traffic volume of 2400 v/h, values of indexes: FC = 0.6; YWC = 2.1. − WII — suburban settlement, passenger traffic − WIIa — a two-way street with the traffic volume of 250 v/h, values of indexes: FC = 0.5; YWC = 1.7 − WIIb — a two-way street with the traffic volume of 600 v/h, values of indexes: FC = 0.4; YWC = 1.5 − WIII — urban settlement, passenger traffic with a 15% share of lightduty trucks − WIIIa — a four-line street with the traffic volume of 1800 v/h, values of indexes: FC = 0.3; YWC = 1.2 − WIIIb — a two-way street with the traffic volume of 3000 v/h with a 5% share of heavy-duty trucks, values of indexes: FC = 0.75; YWC = 2.7 − WIV — a fast highway within a city, four-line transit traffic with the traffic volume of 6300 v/h with a 20% share of heavy-duty trucks, values of indexes: FC b 0.2; YWC = 1.0. Vehicle traffic volume was averaged for the time interval from 7 A.M. to 7 P.M.

TAQI (mg km-1s-1)

QdL·Ywc(kW km-1)

35000

8000

30000

7000

25000

6000 5000

20000

4000

15000

3000

10000

2000

5000

1000

0

0 WIa

WIb

WIIa

WIIb

WIIIa

Fig. 5. TAQI values — winter.

WIIIb

WIV

WIa

WIb

WIIa

WIIb

WIIIa

WIIIb

Fig. 6. Corrected heat emission fluxes QdL · YWC — summer.

WIV

612

Z. Bagieński / Science of the Total Environment 505 (2015) 606–614

QdL (kW km-1) 8000 7000 6000 5000 4000 3000 2000 1000 0 WIa

WIb

WIIa

WIIb

WIIIa

WIIIb

WIV

Fig. 7. Heat emission fluxes QdL — winter.

3.1.1. Vehicle characteristics Table 3 shows a percentage of vehicle types in car traffic scenarios. The following was defined based on Poland's 2010 statistical data for particular traffic variations: vehicle categories (engine capacity), kinds of fuel (petrol, oil, LPG) and the age of a vehicle, which determine which EURO standards need to be conformed to. 3.1.2. Vehicle velocity characteristics The following average velocity values for urban vehicles were accepted to calculate emission following EMEP/EEA guidelines and field measurements: WI w = 25 km/h; WII w = 50 km/h; WIII w = 40 km/h; WIV w = 70 km/h. 3.1.3. Time intervals for which calculations were made − Summer period: averaging Poznań climatic conditions for June, July, August; t = 18.0 °C, relative humidity 45%. − Winter period: averaging climatic conditions for December, January, February; t = 0.5 °C, relative humidity 90%. The example considers vapor emissions only related to engines. 3.2. TAQI value calculation results for selected streets Tables 4 and 5 show fluxes of emission of particular pollutants, flux of an equivalent emission and the TAQI index for summer and winter periods for the analyzed streets. Figs. 2 and 3 show flux of an equivalent

Fig. 9. Niestachowska street — analyzed street section.

emission values EtR for streets during summer and winter. Car traffic volume and kinds of vehicles, especially truck transport, determine the emission values of particular substances and equivalent emission. Of course, emission values during winters are higher, which stems from higher fuel consumption and a bigger share of thermodynamically unfavorable fuel combustion conditions. The street canyon structure described by the value of the YWC has a considerable impact on pollution diffusion conditions. A high value of the FC of the canyon results in a significant rise in concentrations of toxic substances, i.e. a deterioration air quality. It is particularly evident for streets: WIa, WIb, and WIIIb, in the cases of which high values of the YWC index, with a moderate value of equivalent emission, led to the WIa and WIb streets being categorized as unhealthy (Category of Air Quality 4) and as hazardous (CAQ 5) in winter. In summer, the streets were assigned equally high CAQ values. Figs. 4 and 5 show the TAQI values for summer and winter. Table 6 shows the values of determinants of UHI being generated by vehicles QdL for the analyzed streets determined according to Eqs. (4) and (5). The values are relatively high for high traffic volume streets. In summer as solar radiation flux density in the UV–VIS range is higher, additional significant anthropogenic heat emission has a considerable impact on chemical reactions that create secondary air pollution. For example Taha (1996) states that an increase in the Los Angeles local air temperature from 22 °C to 32 °C caused a rise in ozone concentration from 0.120 ppm to 0.240 ppm. Such a situation takes place especially in narrow street canyons, or areas with limited ventilation conditions, a low SVF (sky view factor) value and high radiation absorption values. QdL · YWC values in summer consider limited air circulation within a canyon. Figs. 6 and 7 show heat emission fluxes: QdL · YWC for summer and QdL for winter for streets. 4. Evaluating relations between emission and pollution concentration

Fig. 8. Garbary street — analyzed street section.

CO concentration measurements were carried out for two streets: Garbary, which corresponds to WIa, and Niestachowska, i.e. WIV, in the meteorological conditions of summer and winter. Figs. 8 and 9 show the photo of analyzed street sections. CO concentrations were measured using the non-dispersive infrared method with a detection limit of 50 ppb. CO analyzers were located approximately 2 m from the street edge at the level of about 2 m. Measurements were taken on working days between 10 A.M. and 12 noon (moderate traffic volume) and 4 P.M. and 6 P.M. (large traffic volume). Wind velocity did not exceed 1 m/s. Measured CO concentration values were referred to CO emission calculated according to COPERT IV procedure. Emissions of the remaining pollutants and linear density flux of an equivalent pollutant EtR and TAQI were also calculated. It was assumed that for the determined emission technical conditions and the small distance between

Z. Bagieński / Science of the Total Environment 505 (2015) 606–614

613

Table 7 CO concentration measurements and analysis of the results — Garbary street. Temp.

Humid.

Time

N

LDT

CCOb

CCO

ECO

CCO/ECO

EtR

TAQI

CtR

C

%

Hours

v/h

%

μg m−3

μg m−3

μg m−1 s−1

s m−2

μg m−1 s−1

μg m−1 s−1

μg m−3

21.5 26.5 20.0 23.5 1.2 −0.5 2.5 0.5

42 34 42 38 85 90 78 85

10–12 4–6 P.M. 10–12 4–6 P.M. 10–12 4–6 P.M. 10–12 4–6 P.M.

1310 1870 1230 1690 1405 1780 1520 1730

8 10 7 12 10 9 6 8

1400 2100 1100 2300 2550 3500 2270 3450

1250 1950 950 2150 2300 3250 2020 3200

2355 3361 2005 3190 4577 5799 4120 5520

0.53 0.58 0.47 0.67 0.50 0.56 0.49 0.58

3820 5453 3587 4928 5665 7178 6129 6976

11,460 16,358 10,760 14,784 16,996 21,533 18,388 20,928

2028 3163 1699 3321 2847 4023 3005 4044

the emission source and the analyzer, the concentration of the pollutant is proportional to its emission. It was therefore possible to estimate the concentration values of other pollutants based on the emission value. Berkowicz et al. (2006) pays attention to significant differences in traffic emission values specified according to various computational methods. As pointed out in Section 2.2, the values of pollution emission indicators according to EMEP/EEA guidelines may differ considerably depending on the kind of fuel, engine capacity and the age of a vehicle (EURO standard), the type of traffic and – especially for CO and NMVOC – meteorological conditions. Therefore, inaccuracies in calculating emission may largely stem from the conditions having been defined improperly. Tables 7 and 8 present the meteorological conditions of the measurements, traffic volume N, share of light-duty trucks (LDT) or heavyduty trucks (HDT), the results of measuring the concentrations of CCOb and computational analysis of the results comprising: CCO refers to concentrations of carbon monoxide after deducting CO background and ECO — the calculated emission linear density flux of carbon monoxide. The value of TAQI considers the factor of YWC for the street. The values of the equivalent concentrations are calculated according to: C tR ¼ EtR  ðC CO =ECO Þ:

ð8Þ

It is the equivalent concentration of SO2 determined on the basis of degrees of harmfulness of emitted pollutants shown in Table 1. Comparing the values of CtR to the standards of SO2 concentrations specified according to NAAQS — US EPA and Air Quality Index levels according to EPA (Air Quality Index, 2003) made it possible to verify the dependence of the air quality category on the TAQI value shown in Table 1. NAAQS — US EPA determines the standard for sulfur oxides for averaging times of 3 h at the level of 1300 μg/m3. Air Quality Index levels for SO2 for averaging time of 1 h are: very unhealthy: 855–1710 μg/m3 and hazardous: 1710–2850 μg/m3. Fig. 10 shows a relation between the linear density flux of an equivalent pollutant emitted from traffic sources EtR and carbon monoxide CCO concentrations for Garbary and Niestachowska streets for real conditions. The relatively high CO concentrations and the high sensitivity to flux pollutant EtR in Garbary street as compared to Niestachowska

street stem from not only the depth of the canyon, but also a share of cold emission and smaller vehicle velocity. 5. Conclusions The traffic air quality index is determined based on the real emission from traffic sources and on the structure of streets and of the buildings near streets. Therefore, it is a useful tool for objective assessment of air quality and potential threat to human health near roadways, especially in compact settlement city sectors. It is important to create an air quality index for forecasting potential threats and act preemptively by influencing the source of emission as a basic anthropogenic factor that has a dynamic impact on air quality, and as well as on the urban structure as an important (anthropogenic) factor that has an impact on pollutant dispersion and on air quality. Understanding the impact of development structure on traffic pollution dispersion conditions is a very important issue in urban planning and air quality management. It is an extensive problem which concerns both existing urban areas, e.g. many old European cities, and also newly-built housing estates and it is not fully understood yet. The paper is supposed to provide a tool to assess a potential threat to health of inhabitants living in a particular urban structure posed by pollutants from traffic sources. The analysis conducted in the paper confirms the considerable impact of street canyon structure on the TAQI values, and hence on the level of threat to the health of inhabitants. It means that in the case of compact urban settlement (e.g. old European cities) attempts must be made to limit or eliminate traffic, especially that of trucks. Transit traffic should be eliminated from built-up areas. If a good indoor air quality (IAQ) level in buildings is merely ensured through air filtration and ventilation system, this may lead to the sick building syndrome (SBS). When new housing estates are planned, a compromise is required between an attempt to obtain compact settlement (which influences costs of communal infrastructure and communication and urban order) and to ensure good ventilation in a street. When developing the methodology of TAQI calculation a series of simplifying assumptions concerning street canyon structure, wind direction and vertical gradients of pollutant concentration, a fluid flow of traffic and pollution dispersion between streets were adopted. In some cases, these assumptions may lead to either an overestimation

Table 8 CO concentration measurements and analysis of the results — Niestachowska street. Temp.

Humid.

Time

N

L/H-DT

CCOb

CCO

ECO

CCO/ECO

EtR

TAQI

CtR

C

%

Hours

v/h

%

μg m−3

μg m−3

μg m−1 s−1

s m−2

μg m−1 s−1

μg m−1 s−1

μg m−3

20.5 22.0 21.5 23.5 2.4 −0.5 1.5 −2.0

40 41 38 35 85 90 88 92

10–12 4–6 P.M. 10–12 4–6 P.M. 10–12 4–6 P.M. 10–12 4–6 P.M.

4980 6850 5160 7040 5350 6720 4750 5870

22 20 17 18 18 17 16 20

450 550 300 600 570 800 450 730

350 450 200 500 420 650 300 580

2590 3526 2180 3420 3180 3910 2590 3856

0.14 0.13 0.09 0.15 0.13 0.17 0.12 0.15

15,176 20,875 15,725 21,454 16,853 21,169 14,963 18,491

15,176 20,875 15,725 21,454 16,853 21,169 14,963 18,491

2051 2664 1443 3137 2226 3519 1733 2782

614

Z. Bagieński / Science of the Total Environment 505 (2015) 606–614

CCO (µg/m-3) 4000 3500 3000 2500 2000 1500 1000 500 0 0

5000

10000

15000

20000

25000

EtR (µg/m-1s-1) Fig. 10. Relation between the flux of an equivalent pollutant EtR and concentrations CCO for Garbary and Niestachowska streets.

or an underestimation of health risk. I attempt to find a compromise between the universality and practical application of the method and its accuracy. Therefore, the methodology concerns determining a potential threat to human health near roadways. The methodology can be used in an urban area with no extreme topographic conditions, e.g. lying close to big water reservoirs, deserts, and mountainous ranges, in the case of which topographic and meteorological conditions often have a decisive impact on air quality. Acknowledgments This research was supported by the internal grant of the Poznan University of Technology #01/13-DSPB/0759. References Ahmad K, Khare M, Chaudhry K. Wind tunnel simulation studies on dispersion at urban street canyons and intersections — a review. J Wind Eng Ind Aerodyn 2005;93: 697–717. Air Quality Index. A guide to air quality and your health; 2003 (EPA-454/K-03-002). Air quality in Europe. EEA technical report; 2013. Bagieński Z. The analysis of dispersion of pollutants from short point sources — wind tunnel experimental investigation. Environ Prot Eng 2006;32(4):37–45. Bagieński Z. Emission from stationary combustion sources as the determinant of energy air quality index. Environ Prot Eng 2011;37(1):39–49. Baldauf RW, Heist D, Isakov V, Perry S, Hagler G, Kimbrough S, et al. Air quality variability near a highway in a complex urban environment. Atmos Environ 2013;64:169–78. Berkowicz R, Winther M, Ketzel M. Traffic pollution modelling and emission data. Environ Model Software 2006;21:454–60.

Carpentieri M, Hayden P, Robin AG. Wind tunnel measurements of pollutant turbulent fluxes in urban intersections. Atmos Environ 2012;46:669–74. Chana TL, Luoa DD, Cheunga CS, Chanb CK. Large eddy simulation of flow structures and pollutant dispersion in the near-wake region of the studied ground vehicle for different driving conditions. Atmos Environ 2008;42:5317–39. Chang Cheng-Hsin, Meroney RN. Concentration and flow distributions in urban street canyons: wind tunnel and computational data. J Wind Eng Ind Aerodyn 2003;91: 1141–54. Clements AL, Jia Y, Denbleyker A, McDonald-Buller E, Fraser MP, Allen DT, et al. Air pollutant concentrations near three Texas roadways, part II: chemical characterization and transformation of pollutants. Atmos Environ 2009;43:4523–34. Crutzen PJ. New directions: the growing urban heat and pollution island effect — impact on chemistry and climate. Atmos Environ 2004;38:3539–40. Directive. Directive 2008/50/EC of the European Parliament and of the Council of 21 May 2008 on ambient air quality and cleaner air for Europe; 2008. EMEP/EEA,2010. Lead authors: Ntziachristos, L., Samaras, Z, Emission inventory guidebook 2009, updated June 2010, European Environment Agency. Environment Canada. http://www.ec.gc.ca, 2013. Fenger J. Urban air quality. Atmos Environ 1999;33:4877–900. Guide to UK Air Pollution Information Resources, 2013. Department of Environment, Food and Rural Affairs. http://www.gov.uk/defra. Guinee JB, editor. Handbook on Life Cycle Assessment. Dordrecht: Kluwer Academic Publishers; 2001. Hanna SR, Britter R, Franzese P. A baseline urban dispersion model evaluated with Salt Lake City and Los Angeles tracer date. Atmos Environ 2003;37:5069–82. Kaster-Klein P, Berkowicz R, Plate EJ. Modeling of vehicle inducted turbulence in air pollution studies for street. Int J Environ Pollut 2000;14:496–507. Kim Y, Guldmann JM. Impact of traffic flows and wind directions on the pollution concentrations in Seoul, Korea. Atmos Environ 2011;45:2803–10. Kim MJ, Park RJ, Kim J-J. Urban air quality modeling with full O3–NOx–VOC chemistry: implications for O3 and PM air quality in a street canyon. Atmos Environ 2012;47: 330–40. Oke TR. Street design and urban canopy layer climate. Energy Build 1988;11:103–13. Park SK, Kim SD, Lee Heekwan. Dispersion characteristics of vehicle emission in an urban street canyon. Sci Total Environ 2004;323:263–71. Peterka JA, Meroney RN, Kothari KM. Wind flow patterns about buildings. J Wind Eng and Ind Aerodyn 1985;21:21–38. Qin Y, Kot SC. Dispersion of vehicular emission in street canyons, Guangzhou city, South China. Atmos Environ 1993;27B:283–91. Regulation of the Minister of Environment (Poland). Regulation of the Minister of Environment (Poland) of 13 September 2012 on the evaluation of air pollutant levels; 2012 (DzU 2012/1032 (in Polish)). Rodriguez LM, Bieringer PE, Warner T. Urban transport and dispersion model sensitivity to wind direction uncertainty and source location. Atmos Environ 2013;64:25–9. Schembari A, Triguero-Mas M, de Nazelle A, Dadvand P, Vrijheid M, Cirach M, et al. Personal, indoor and outdoor air pollution levels among pregnant women. Atmos Environ 2013;64:287–95. Sini JF, Anquetin S, Mestayer P. Pollutant dispersion and thermal effects in urban street canyons. Atmos Environ 1996;30:2659–77. Sokhi RS. Urban air quality. Atmos Environ 2008;42:3909–10. Taha H. Modeling the impacts of increased Urban vegetation on the ozone air quality in the south coast air basin. Atmos Environ 1996;30:3423–33. TREMOVE. Service contract for the further development and application of the transport and environmental TREMOVE model Lot 1, 2007. Final Report European Commission, Directorate General Environment, Energy and Environment, 9 July 2007; 2007. WHO. Air quality guidelines for particulate matter, ozone, nitrogen dioxide and sulfur dioxide. Global update 2005; 2006.