Using an innovative flag element ratio approach to tracking potential sources of heavy metals on urban road surfaces

Environmental Pollution 243 (2018) 410e417

Contents lists available at ScienceDirect

Environmental Pollution journal homepage: www.elsevier.com/locate/envpol

Using an innovative flag element ratio approach to tracking potential sources of heavy metals on urban road surfaces* Nian Hong a, b, Panfeng Zhu a, An Liu a, b, c, *, Xu Zhao d, Yuntao Guan e a

College of Chemistry and Environmental Engineering, Shenzhen University, 518060, Shenzhen, China Shenzhen Key Laboratory of Environmental Chemistry and Ecological Remediation, 518060, Shenzhen, China c Science and Engineering Faculty, Queensland University of Technology (QUT), Brisbane, Australia d Key Laboratory of Drinking Water Science and Technology, Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences, Beijing, 100085, China e Guangdong Provincial Engineering Technology Research Centre for Urban Water Cycle and Water Environment Safety, Graduate School at Shenzhen, Tsinghua University, 518055, Shenzhen, China b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 20 June 2018 Received in revised form 15 August 2018 Accepted 30 August 2018 Available online 5 September 2018

Heavy metals deposited on urban road surfaces can be washed-off by stormwater runoff, undermining stormwater reuse safety due to their high toxicity to ecological and human health. Heavy metals on urban road surfaces come from diverse sources and tracking these sources is essential to effectively manage stormwater and hence its reuse safety. This research study developed an innovative approach to tracking sources of heavy metals using data collected in Shenzhen, China. This approach developed was based on a “flag element ratio” theory, where each source generally corresponds to a specific ratio of targeted pollutants to the flag element. It is noted that Cr, Cu, Pb, Ni, and Zn on urban roads were 19.05 mg/kg to 152.01 mg/kg, 25.66 mg/kg to 310.75 mg/kg, 15.61 mg/kg to 220.35 mg/kg, 10.65 mg/kg to 100.28 mg/kg, and 138.14 mg/kg to 1047.05 mg/kg, respectively. Gasoline emission was the main source for Cr, Ni and Pb, while braking wear and tyre wear were the major sources of Cu and Zn, respectively. Furthermore, the rankings of sources of each heavy metal in terms of their contributions were obtained by using this approach. Vehicle exhaust was found as the main contributor for all the heavy metals on urban road surfaces. This highlighted that vehicle exhaust should be seriously considered in terms of controlling heavy metal pollution on urban road surfaces and hence resulting urban road stormwater runoff. © 2018 Elsevier Ltd. All rights reserved.

Keywords: Stormwater reuse Heavy metals Source tracking Urban road surfaces

1. Introduction Stormwater has been considered as an important alternative water resource for mitigating water shortage issues worldwide. As primary impervious surfaces in the urban environment, urban roads can generate a larger amount of stormwater runoff which has high potential to be reused (Al-Salaymeh et al., 2011, Sa'd A et al., 2014). However, a range of toxic pollutants such as heavy metals from traffic activities usually deposit (build-up) on urban road surfaces during dry periods (Zafra et al., 2016). When rainfall event occurs, these pollutants can be washed-off by stormwater runoff

*

This paper has been recommended for acceptance by Joerg Rinklebe. * Corresponding author. College of Chemistry and Environmental Engineering, Shenzhen University, 518060, Shenzhen, China. E-mail address: [email protected] (A. Liu). https://doi.org/10.1016/j.envpol.2018.08.098 0269-7491/© 2018 Elsevier Ltd. All rights reserved.

(Liu et al., 2013, 2016; Zafra et al., 2017), posing high risks to ecological and human health when the stormwater is reused (Liu et al., 2015; Zhao et al., 2011). Generally, pollutants build-up determines the availability for their wash-off prior to rainfall events and hence source characteristics of pollutants build-up might influence pollutant concentrations in wash-off (Zafra et al., 2016, 2017). On urban road surfaces, heavy metals come from diverse sources such as wear of vehicle, engine exhaust, road wear and roadside soil and are primarily attached to road deposited sediments (RDS) which are the mixture of particles (Gunawardena et al., 2015; Mummullage et al., 2016; Zhao et al., 2011). In this context, an indepth understanding of source characteristics in heavy metal build-up on urban road surfaces and identification of the key sources can contribute to heavy metal pollution control and hence ensuring stormwater reuse safety (Zhao and Li, 2013).

N. Hong et al. / Environmental Pollution 243 (2018) 410e417

Several approaches to tracking sources of pollutants in the urban environment have been reported. These approaches primarily include principal component analysis (PCA) (Shi et al., 2011), positive matrix factorization (PMF) (Song et al., 2008; Yatkin and Bayram, 2008), chemical mass balance (CMB) (Liu et al., 2008; Shi et al., 2009) and diagnostic ratios method (Zhang et al., 2008). PCA and PMF are the approach to reduce a set of raw data into a number of principal components, where the pollutant sources are identified accordingly. Many past research studies used PCA and/or PMF to identify sources of heavy metals in the urban environment (Egodawatta et al., 2013; Gunawardena et al., 2014, 2015; Mummullage et al., 2016). Since these two approaches do not include all information of the raw dataset, this could underestimate the number of sources (Zhang et al., 2012). CMB method is mainly applied to identify sources of atmospheric pollutants (Chow and Watson, 2002) while diagnostic ratios method is mainly utilized in polycyclic aromatic hydrocarbons (PAHs) source apportionment studies (Zhang et al., 2008). A flag element ratio approach has been widely used in the pollutant source tracking in the area of groundwater research (Katz et al., 2011; Kelly et al., 2010). In most of previous research studies, chloride in groundwater was tracked using Cl/Br ratio method from various anthropogenic and natural sources. This approach is based on the theory that each source generally corresponds to a specific ratio (such as Cl/Br ratio in the groundwater research) (Panno et al., n-Zapata et al., 2014). Based on the theory above, 2005, 2006; Paste this study extended this flag element ratio method to heavy metal pollution on urban roads. It is hypothesized that each source of heavy metals on urban roads has a specific ratio of targeted heavy metal to selected flag element. Using the flag element ratio method to tracking sources of heavy metals on urban roads has not been done in previous stormwater research studies. Unlike PCA, PMF and CMB methods, the flag element ratio method does not require complex parameters and mathematical calculations and hence would be relatively easily extended to other fields such as receiving water pollution source tracking and checking whether some possible sources significantly contribute to the targeted environmental pollution. The knowledge developed is expected to contribute to effective pollutant source control strategy implementation on urban roads and hence ensure urban road stormwater reuse safety. It also showcased the extended application of the source tracking method in different domains of environmental pollution research. 2. Methods and materials 2.1. Source tracking approach development In flag element ratio method, a major element (flag element) usually has a much higher amount (two or three orders of magnitude higher) than those pollutants which are targeted for source tracking in the natural environment. When various sources start to contribute, the amount ratio (load or concentration) of the flag element and targeted pollutants primarily vary with the amount of targeted pollutants while the ratio is less influenced by the flag element amount due to their much higher amounts. In this study, Iron (Fe) was considered as the flag element. This was due to the fact that Fe is among the most important natural geogenic metals and Fe loads had two or three orders of magnitudes higher than anthropogenic activities related heavy metals (such as Cu, Pb and Zn). This can be evident by several previous studies. For example, Hong et al. (2017) found that the Fe concentrations (representing the Fe loads per unit mass of RDS) on urban roads ranged from 11588 mg/kg to 59550 mg/kg while mean loads of other heavy metals (primarily related to vehicle

411

traffic) were 128.85 ± 83.86 mg/kg (Cu), 400.44 ± 253.95 mg/kg (Zn), 36.74 ± 22.73 mg/kg (Ni), 62.08 ± 32.14 mg/kg (Cr), 59.03 ± 56.12 mg/kg (Pb). Apeagyei et al. (2011) also noted that Fe concentrations attached to solids on urban roads ranged from 10044 to 100077 ppm, which were almost 100 to 1000 times higher than other heavy metals. In order to develop the “flag element ratio” source tracking approach, the following assumptions were applied:
The RDS is a mixture of pollutant source particles, including low heavy metal concentration particles (PLC) and high heavy metal concentration particles (PHC);
For all the RDS samples, the sources were same;
The chemical proprieties of source particles were stable;
No chemical relations during the mixture of sources. Accordingly, in an ideal condition which only includes PLC and PHC, the concentrations of heavy metal element A and Fe can be calculated as follow:

CA ¼

 r r   CH;A þ 1   CL;A 100 100

CFe ¼

 r r   CH;Fe þ 1   CL;Fe 100 100

(1)

(2)

where CA and CFe are concentrations of element A and Fe (mg/kg) in the RDS particles (including PLC and PHC), respectively; r is mixing portion (%) of PHC; CH,A and CH,Fe are concentrations of element A and Fe (mg/kg) in PHC, respectively; and CL,A and CL,Fe are concentrations of element A and Fe (mg/kg) in PLC, respectively. Generally, they are CH,A > CL,A and the Fe-A ratio CH,Fe/CH,A < CL,Fe/CL,A in natural conditions. In Eqs. (1) and (2), CH,A, CH,Fe, CL,A and CL,Fe were constant values. Then CA and CFe were considered as the function of r. The increased value of r (from 0% to 100%) indicates that the pollutant source particles continuously contribute to RDS and hence the A concentrations of RDS increase. This leads to the value of Fe-A ratio (CFe/CA) decreasing since Fe has much higher amounts than anthropogenic activities related heavy metals. The theoretical calculation curves of Fe-A ratio versus element A concentration with the increased value of r from 0% to 100% is illustrated in Fig. 2. According to previous studies, PHC generally account for a small portion of RDS (Hong et al., 2017; Liu et al., 2015). Namely, r is usually a small value (usually less than 40%) in the RDS particles. According to Fig. 1, the curves of Fe-A ratio versus element A concentration in logarithm can be roughly regarded as a straight line under the condition of r in small percentage values (shown using the red direction arrow in Fig. 1). In the case of r in larger percentage values, there are three possible conditions theoretically: (1) the slope of the curves decreased with increased A concentrations (Fig. 1a); (2) the slope of the curves keep constant with increased A concentrations (Fig. 1b); (3) the slope of the curves increased with increased A concentrations (Fig. 1c). The red direction arrow generally points to the primary sources. Fig. 2 shows the conceptual figure of heavy metal source identification. The potential source zone (red square area) is formed by the point of direction arrow and the horizontal boundary (reflecting slope decreasing trend as shown in Fig. 1a) and vertical boundary (reflecting slope increasing trend as shown in Fig. 1c). In PHC, the larger overlapped area between distribution area (dot line circle in Fig. 2) and potential source zone, the higher potential to become the primary source. Considering more than one PHC in the real world, the degree of overlapped area between PHC distribution area and potential source zone can be considered as criteria to rank

412

N. Hong et al. / Environmental Pollution 243 (2018) 410e417

Fig. 1. Three possible tendencies for the curves of Fe-A ratio versus heavy metal A concentration. (PLC: low heavy metal concentration particles; PHC: high heavy metal concentration particles; r: mixing portion (%) of PHC in road deposited sediment).

Fig. 2. A conceptual figure of source identification for heavy metals on urban road surfaces using the Fe-A ratio approach (RDS: road deposited sediment; PLC: low heavy metal concentration particles; PHC 1, 2, and 3: high heavy metal concentration particles 1, 2 and 3 which completely, partly and not falling into the potential source zone, respectively).

sources’ contribution. For example, the PHCs completely falling into the potential source zone (namely the area of the red direction arrow pointing, refer to PHC 1 in Fig. 2) are considered as the primary sources of the heavy metal A. The sources partly falling into the red square area (refer to PHC 2 in Fig. 2) were considered as the secondary sources of the heavy metal A while the sources not falling into the red square area (refer to PHC 3 in Fig. 2) are considered as not important sources of the heavy metal A.

2.2. Sample collection The samples in the study had two groups. One group was the pollutant source particle samples and another group was RDS samples. These two groups of samples were collected in Shenzhen, China, which was recognized as a typical megacity located at the north coast of South China Sea. The average annual rainfall is 1917 mm and more than 80% rainfall occurs from April to September, the average annual temperature is 23.3  C (Meteorological Bureau of Shenzhen Municipality, 2018).

2.2.1. Pollutant source particle sample collection In terms of source particle samples, eight types of possible source particles were collected. They were brake lining dusts (LD), brake shoe dusts (SD), gasoline engine exhaust (GE), diesel engine exhaust (DE), road asphalt particles (AS), road aggregate particles (AG), tyre wear (TW) and urban soil (US). These were identified by a number of previous research studies as the most important sources of heavy metals on urban road surfaces (Thorpe and Harrison, 2008; Vouitsis et al., 2009; Wang et al., 2003; Zhao et al., 2011; Adachi and Tainosho, 2004; Chang et al., 2009). It is noteworthy that other than these eight sources, there are still several other possible sources contributing to heavy metal pollution on urban roads such as atmospheric deposition. This study was to use the eight primary sources to showcase the development of the flag element ratio source tracking method and more source types would be included in other cases where are applicable. LD and SD were collected from the vehicle braking systems while data relevant to TW was obtained from a previous study (Kennedy and Gadd, 2000). GE and DE were collected from the exhaust pipes of gasoline and diesel vehicles respectively. AS and AG are the components of urban road pavements and collected from the urban road surfaces. US was collected at the green belts of urban roads. The detailed information about the pollutant source particle sample collection is given in Table 1.

2.2.2. RDS sample collection For RDS sample collection, twenty road sites were selected in Shenzhen. All the sampling road sites are relatively flat and have curbs at both sides. All road surfaces were paved with asphalt. These sampling road sites encompassing different land use types (residential, commercial and industrial), traffic lanes (2e14 lanes) and traffic volumes (868e50960 vehicle/day), which can be considered as representative urban road sites for RDS collection. RDS samples were collected using dry and wet vacuuming method (Egodawatta et al., 2009; Gunawardana et al., 2013; Mahbub et al., 2011a; b). During the sample collection procedure, the road surfaces were vacuumed in dry condition and then in wet condition (without creating runoff) for collecting both particulate and dissolved pollutant on a certain test plot (2 m  2 m) of each road site. The method using a water filter contained vacuum cleaner (Haier, ZTBJ1200, China, Power: 1200 W) to collect pollutant, a sprayer to wet the test plot of each road site and a frame to demarcate the test plot. In this study, the antecedent dry period was seven days before RDS sample collection because pollutant build-up on road surfaces

N. Hong et al. / Environmental Pollution 243 (2018) 410e417

413

Table 1 Information of pollutant source particle sample collection. Source particle samples

No. of Sampling method samples

Brake lining dust (LD) 32

The wear particles were collected from waste brake linings.

Brake shoe dust (SD)

8

The wear particles were collected from waste brake shoes.

Gasoline engine exhaust particle (GE)

25

The exhaust particles were collected from vehicle exhaust pipes

Photo

Other information The waste brake lining and brake shoe were collected from a vehicle repairing plant in the same study area.

Collected from a gasoline engine vehicle

Collected from a diesel engine vehicle

Diesel engine exhaust 9 particle (DE)

Road asphalt particle (AS)

12

The road surface sample was smashed and then isolated into AS and AG.

Road aggregate particle (AG)

12

Urban soil (US)

12

The US samples were collected at 0.3 m below the surface of green belts

Tyre wear particle (TW)

14

Data relevant to tyre wear particles were obtained from a previous study

almost became a constant value after a seven-day antecedent dry period (Egodawatta, 2007). Accordingly, a total of 20 RDS samples were collected from the 20 road sites.

2.3. Laboratory testing RDS and pollutant source particle samples were tested for chromium (Cr), copper (Cu), nickel (Ni), lead (Pb), zinc (Zn) and iron (Fe). Cr is considered as a highly toxic heavy metal element for ecological and human health (Bahrami et al., 2016); Cu and Ni are not only important metal element for plating and alloy materials, but also are primary heavy metal types in urban road stormwater pollution (Thorpe and Harrison, 2008); Pb is a priority controlled heavy metal species in the urban environment (Vile et al., 2000) while Zn is found to have a relatively higher build-up loads on road surfaces than other heavy metals and hence higher concentrations

Refer to Kennedy and Gadd (2000)

in urban stormwater runoff (Ma et al., 2017). Fe is the flag element which was used for identifying the sources of other heavy metals. The samples including RDS, LD, SD, GE, DE, AG and US were firstly digested using mix acid according to Method 3030H (APAH, 2005). The AS samples were firstly ashed by a muffle furnace using the following parameters: 115  C for 3 h; 130  C for 2 h; 150  C for 1 h; 250  C for 1 h; 350  C for 1 h; 400  C for 4.5 h (Kennedy and Gadd, 2000). Then the ashed AS samples were digested using the same method as above. Then, all of the digested samples were tested according to Method 3120 (APAH, 2005) by Inductively Coupled Plasma-Optical Emission Spectrometer (ICP-OES) (Optima 7300 DV, Perkin-Elmer, USA). The recovery ranged from 75.6% to 102.3%, which is within the acceptable range of 75%e120% (Herngren et al., 2005). In addition, the total solids were also tested by the gravimetric method according to Methods 2540C and 2540D (APAH, 2005).

414

N. Hong et al. / Environmental Pollution 243 (2018) 410e417

3. Results and discussion 3.1. Heavy metal concentrations of pollutant source and RDS particles The comparison of heavy metal concentrations in each source and RDS samples are shown in Fig. 3. In terms of RDS (shown as black boxes in Fig. 3), the concentrations of Cr, Cu, Pb, Ni, and Zn were 19.05 mg/kg to 152.01 mg/kg, 25.66 mg/kg to 310.75 mg/kg, 15.61 mg/kg to 220.35 mg/kg, 10.65 mg/kg to 100.28 mg/kg, and 138.14 mg/kg to 1047.05 mg/kg, respectively. In terms of pollutant source particles, they are generally divided into two groups based on their heavy metal concentrations comparing to RDS. AG, AS, US and TW can be considered as PLC group (shown as blue boxes in Fig. 3) which had relatively lower mean values of concentration compared to RDS for all heavy metals (except that Zn had a high concentration in TW particle samples) while SD, LD, GE and DE can be seen as PHC group (shown as red boxes in Fig. 3) which had higher mean values of concentration compared to RDS. The heavy metal concentrations of RDS (shown as black boxes in Fig. 3) were generally between the low (AG, AS, US and TW) and

high (SD, LD, GE and DE) heavy metal concentration groups. These observations mean that vehicle braking wear and exhaust emission significantly contribute to heavy metals on urban road surfaces while road surface pavement materials and soils are less contributing. However, the tyre wear could be the important source of Zn on urban surfaces. This was due to the fact that ZnO is added as an activator during the tyre vulcanisation process (Councell et al., 2004; Kennedy and Gadd, 2000; Smolders and Degryse, 2002). 3.2. Source tracking of heavy metals for RDS Fig. 4 shows the source tracking of Cr, Cu, Ni, Pb, and Zn using the flag element (Fe in the study) ratio approach discussed in Section 2.1. For all the heavy metal species, RDS samples generally go along the red direction arrows and point to the red square areas which represent the primary sources (see Section 2.1). Additionally, the Fe-heavy metal ratios decreased with the increased heavy metal concentrations in terms of all heavy metals. It is noted that pollutant source particles which generally fell into the red square areas are GE and DE (Cr), GE and LD (Cu), GE (Ni), GE (Pb) as well as TW and GE (Zn). These results indicate the important role of gasoline emission on contributing to heavy metals on urban road

Fig. 3. Heavy metal concentrations for pollutant sources and RDS particles (The concentration axes are in logarithmic forms; AG: road aggregate AS: road asphalt US: urban soil SD: bake shoe dust LD: bake lining dust GE: gasoline engine exhaust DE: diesel engine exhaust TW: tyre wear RDS: road deposit sediment).

N. Hong et al. / Environmental Pollution 243 (2018) 410e417

415

Fig. 4. Source tracking of Cr, Cu, Ni, Pb and Zn on urban road surfaces (AG: road aggregate AS: road asphalt US: urban soil SD: bake shoe dust LD: bake lining dust GE: gasoline engine exhaust DE: diesel engine exhaust TW: tyre wear RDS: road deposit sediment).

surfaces since GE samples fell into the red square areas for all heavy metals. Furthermore, diesel emission, brake lining dusts and tyre wear also contribute to Cr, Cu and Zn respectively. Previous studies (Egodawatta et al., 2009; Gunawardena et al., 2014, 2015; Mummullage et al., 2016) indicated that Pb came from geogenic sources and road side soil. However, this study showed that Pb concentrations in AG, AS and US (PLC) obviously lower that RDS particles. This means that these sources provide few contributions for Pb pollution in RDS. In addition, during the factor identification process of PCA method in these previous studies, Pb was identified from non-vehicle emission sources due to the utilization of unleaded fuel at present. For examples, Pb is used in battery materials and electric systems of vehicle components (Pichtel et al., 2000; Soundarrajan et al., 2012). However, although Pb has been forbidden for fuel additive, Pb still showed relatively high concentrations in vehicle exhaust particles in our measurements and other research studies (Johansson et al., 2009; Kummer et al., 2009). In this context, vehicle exhaust still should be

considered as a potential main contributor for Pb pollution on urban road surfaces. Therefore, this implies that the flag element ratio approach which contains various source properties could be more reasonable and reliable for source tracking of heavy metal on urban road surfaces than other approaches such as PCA and PMF. 3.3. Ranking of sources’ contribution As shown in Section 2.1, the pollutant source particle samples which completely fall into the red square areas (refer to Fig. 3) can be considered as the most important sources, followed by those source samples partly falling into the red square areas. The sources which do not fall into the red square areas can not be considered as the important sources. Taking Cr as an example (see Fig.4a), all of GE samples were located in the red square areas while DE samples partly fall into the area. In addition, a few LD and SD samples were also located at the red square areas. Therefore, it could be concluded that the ranking of sources' contribution to Cr on urban

416

N. Hong et al. / Environmental Pollution 243 (2018) 410e417

road surfaces is in the order of GE > DE > LD z SD while other sources do not play an important role in contributing Cr since they do not fall into the red square area. The similar approach was applied for other heavy metals in terms of ranking their sources’ contribution and the ranking results are given in Table 2. According to Table 2, GE generally had the highest (þþþ) or the second highest (þþ) ranking for all the heavy metal species while DE had a certain ranking value (þþ or þ) on all heavy metals. LD had ranking values on Cr, Cu, Pb and Zn. TW and SD only had the ranking values on Zn and Cr respectively. This means that vehicle exhaust emission (gasoline and diesel) should be preferentially targeted in terms of mitigating heavy metal pollution in the urban road stormwater. Sources related to braking wear (brake shoe dusts and brake lining dusts) also need to be focused while tyre wear should be given a particular attention for Zn pollution.

for future practical applications. Firstly, this study only included eight source types. More source types such as atmospheric deposition might be added to improve the source tracking method. Secondly, sources’ contributions in this study were roughly ranked by the degree of overlapped area (completely, partly and not overlapped) between distribution area and potential source zone (See Fig. 2). This ranking method is qualitative. In this context, the quantitative ranking method needs to be developed in the future research. Thirdly, this study only investigated five heavy metals (Cr, Cu, Ni, Pb and Zn). However, heavy metals in the urban environment are much more than those. Other heavy metals such as Cd, Hg and As should be added in future research studies in order to have a more comprehensive understanding of heavy metal pollution in the urban environment. 4. Conclusions

3.4. Practical applications of the source tracking approach developed In this study, the approach developed was used to track sources of heavy metals on urban road surfaces. It involves five steps: (1) collection of source samples (the eight types of source samples in this study) and targeted substance samples (RDS samples in this study); (2) testing of targeted pollutants (heavy metals in this study); (3) identification of the flag element (Fe in this study); (4) plotting the figures of the flag element-A ratio versus targeted pollutant A (such as Fig.4 in this study); and (5) identification and ranking of sources contributing to targeted pollutants (such as Fig.4 and Table 2 in this study). In fact, the approach developed can be also used in other fields of pollutant source tracking by using the five steps above. For example, in order to track sources of heavy metals in the atmospheric environment, possible source samples (such as vehicle exhaust emission, emission from surrounding industrial factories and soil samples which can enter the atmosphere by wind and traffic turbulence) as well as air samples can be collected. By plotting the flag element ratio figures (similar as Fig.4 in this study), the primary sources of air pollution by heavy metals can be identified. Additionally, it is likely to rank these sources based on their contributions. Furthermore, this approach can be also applicable to check whether some sources significantly contribute to the targeted environmental pollution. For instance, it is unknown whether sources such as wastewater drainage contribute to the nitrogen pollution of a receiving water body. Collecting wastewater samples and the receiving water samples, identifying a flag element and then plotting the flag element ratio figures might help to find out the answers. Although the developed method is useful to track sources of heavy metals on urban roads, there are still improvements needed

Table 2 Ranking of sources’ contribution to heavy metals on urban road surfaces.

Cr Cu Ni Pb Zn

AG(1)

AS

US

SD

LD

GE

DE

TW

/(3) / / / /

/ / / / /

/ / / / /

þ(2) / / / /

þ þþþ / þ þ

þþþ þþ þþþ þþþ þþ

þþ þ þ þ þ

/ / / / þþþ

Note (1): AG ¼ road aggregate, AS ¼ road asphalt, US ¼ urban soil, SD ¼ brake shoe dust, LD ¼ brake lining dust, GE ¼ gasoline engine exhaust, DE ¼ diesel engine exhaust, TW ¼ tyre wear. (2): þ ~þþþ ¼ the lowest ranking ~ the highest ranking of sources' contribution. (3): the symbol/represents sources which do not significantly contribute to heavy metals on urban road surfaces.

An innovative approach called ‘flag element ratio’ was developed to tracking sources of heavy metals on urban road surfaces. The approach developed is possible to be applied in other fields such as receiving water pollution source tracking and checking whether some possible sources significantly contribute to the targeted environmental pollution. Although this source tracking method still has a few limitations, it can provide useful information regarding source identification and ranking. In terms of source tracking for heavy metals on road surfaces, gasoline emission was identified as the main source for Cr, Ni and Pb, while braking wear and tyre wear were the major sources of Cu and Zn, respectively. Vehicle exhaust was found as the main contributor for all the heavy metals on urban road surfaces. This highlighted that vehicle exhaust should be seriously considered in terms of controlling heavy metal pollution on urban road surfaces and hence resulting urban road stormwater runoff. These results can provide useful insight to ensuing stormwater reuse safety. Acknowledgment We thank National Natural Science Foundation of China [grant number 41601510, 21806110], Shenzhen Science and Innovation Commission [grant numbers ZDSYS201606061530079, JSGG20170412145935322], China Postdoctoral Science Foundation [grant number 2017M622791] and the Development and Reform Commission of Shenzhen Municipality (urban water recycling and environment safety program) to support this research study. Appendix A. Supplementary data Supplementary data related to this article can be found at https://doi.org/10.1016/j.envpol.2018.08.098. References Sa’d A, S., Jaber, F.H., Lesikar, B.J., 2014. Modeling the effect of cistern size, soil type, and irrigation scheduling on rainwater harvesting as a stormwater control measure. Water Resour. Manag. 28 (12), 4219e4235. Adachi, K., Tainosho, Y., 2004. Characterization of heavy metal particles embedded in tire dust. Environ. Int. 30 (8), 1009e1017. Al-Salaymeh, A., Al-Khatib, I.A., Arafat, H.A., 2011. Towards sustainable water quality: management of rainwater harvesting cisterns in southern Palestine. Water Resour. Manag 25 (6), 1721e1736. APAH, 2005. Standard Methods for the Examination of Water and Wastewater 20th Edition. American Public Health Association, American Water Works Association and Water Environment Federation, Washington DC. Apeagyei, E., Bank, M.S., Spengler, J.D., 2011. Distribution of heavy metals in road dust along an urban-rural gradient in Massachusetts. Atmos. Environ. 45 (13), 2310e2323. Bahrami, M., Heidari, M., Ghorbani, H., 2016. Variation in antioxidant enzyme activities, growth and some physiological parameters of bitter melon (Momordica charantia) under salinity and chromium stress. J. Environ. Biol. 37 (4), 529e535.

N. Hong et al. / Environmental Pollution 243 (2018) 410e417 Chang, S.H., Wang, K.S., Chang, H.F., Ni, W.W., Wu, B.J., Wong, R.H., Lee, H.S., 2009. Comparison of source identification of metals in road-dust and soil. Soil Sediment Contam. Int. J. 18 (5), 669e683. Chow, J.C., Watson, J.G., 2002. Review of PM2. 5 and PM10 apportionment for fossil fuel combustion and other sources by the chemical mass balance receptor model. Energy Fuel. 16 (2), 222e260. Councell, T.B., Duckenfield, K.U., Landa, E.R., Callender, E., 2004. Tire-wear particles as a source of zinc to the environment. Environ. Sci. Technol. 38 (15), 4206e4214. Egodawatta, P., 2007. Translation of Small-plot Scale Pollutant Build-up and Washoff Measurements to Urban Catchment Scale. PhD Dissertation. Queensland University of Technology, Brisbane. Egodawatta, P., Thomas, E., Goonetilleke, A., 2009. Understanding the physical processes of pollutant build-up and wash-off on roof surfaces. Sci. Total Environ. 407 (6), 1834e1841. Egodawatta, P., Ziyath, A.M., Goonetilleke, A., 2013. Characterising metal build-up on urban road surfaces. Environ. Pollut. 176, 87e91. Gunawardana, C., Goonetilleke, A., Egodawatta, P., 2013. Adsorption of heavy metals by road deposited solids. Water Sci. Technol. 67 (11), 2622e2629. Gunawardena, J., Ziyath, A.M., Egodawatta, P., Ayoko, G.A., Goonetilleke, A., 2014. Mathematical relationships for metal build-up on urban road surfaces based on traffic and land use characteristics. Chemosphere 99, 267e271. Gunawardena, J., Ziyath, A.M., Egodawatta, P., Ayoko, G.A., Goonetilleke, A., 2015. Sources and transport pathways of common heavy metals to urban road surfaces. Ecol. Eng. 77, 98e102. Herngren, L., Goonetilleke, A., Ayoko, G.A., 2005. Understanding heavy metal and suspended solids relationships in urban stormwater using simulated rainfall. J. Environ. Manag. 76 (2), 149e158. Hong, N., Zhu, P., Liu, A., 2017. Modelling heavy metals build-up on urban road surfaces for effective stormwater reuse strategy implementation. Environ. Pollut. 231, 821e828. Johansson, C., Norman, M., Burman, L., 2009. Road traffic emission factors for heavy metals. Atmos. Environ. 43 (31), 4681e4688. Katz, B.G., Eberts, S.M., Kauffman, L.J., 2011. Using Cl/Br ratios and other indicators to assess potential impacts on groundwater quality from septic systems: a review and examples from principal aquifers in the United States. J. Hydrol. 397 (3e4), 151e166. Kelly, W.R., Panno, S.V., Hackley, K.C., Hwang, H.H., Martinsek, A.T., Markus, M., 2010. Using chloride and other ions to trace sewage and road salt in the Illinois Waterway. Appl. Geochem. 25 (5), 661e673. Kennedy, P., Gadd, J., 2000. Preliminary Examination of Trace Elements in Tyres, Brake Pads, and Road Bitumen in New Zealand. Revised 2003. Kingett Mitchell Ltd. Kummer, U., Pacyna, J., Pacyna, E., Friedrich, R., 2009. Assessment of heavy metal releases from the use phase of road transport in Europe. Atmos. Environ. 43 (3), 640e647. Liu, Y., Shao, M., Lu, S., Chang, C.C., Wang, J.L., Fu, L., 2008. Source apportionment of ambient volatile organic compounds in the Pearl River Delta, China: Part II. Atmos. Environ. 42 (25), 6261e6274. Liu, A., Egodawatta, P., Guan, Y.T., Goonetilleke, A., 2013. Influence of rainfall and catchment characteristics on urban stormwater quality. Sci. Total Environ. 444, 255e262. Liu, A., Liu, L., Li, D.Z., Guan, Y.T., 2015. Characterizing heavy metal build-up on urban road surfaces: implication for stormwater reuse. Sci. Total Environ. 515, 20e29. Liu, A., Gunawardana, C., Gunawardena, J., Egodawatta, P., Ayoko, G.A., Goonetilleke, A., 2016. Taxonomy of factors which influence heavy metal buildup on urban road surfaces. J. Hazard Mater. 310, 20e29. Ma, Y.K., Liu, A., Egodawatta, P., McGree, J., Goonetilleke, A., 2017. Assessment and management of human health risk from toxic metals and polycyclic aromatic hydrocarbons in urban stormwater arising from anthropogenic activities and traffic congestion. Sci. Total Environ. 579, 202e211. Mahbub, P., Goonetilleke, A., Ayoko, G.A., 2011a. Prediction model of the buildup of volatile organic compounds on urban roads. Environ. Sci. Technol. 45 (10), 4453e4459. Mahbub, P., Goonetilleke, A., Ayoko, G.A., Egodawatta, P., Yigitcanlar, T., 2011b. Analysis of build-up of heavy metals and volatile organics on urban roads in

417

gold coast, Australia. Water Sci. Technol. 63 (9), 2077e2085. Meteorological Bureau of Shenzhen Municipality, 2018. Shenzhen climate bulletin. Available at: http://www.szmb.gov.cn/qixiangfuwu/qihoufuwu/ qihouguanceyupinggu/nianduqihougongbao/. (Accessed 5 August 2018). Mummullage, S., Egodawatta, P., Ayoko, G.A., Goonetilleke, A., 2016. Use of physicochemical signatures to assess the sources of metals in urban road dust. Sci. Total Environ. 541, 1303e1309. Panno, S.V., Hackley, K.C., Hwang, H.H., Greenberg, S., Krapac, I.G., Landsberger, S., O'Kelly, D.J., 2005. Database for the characterization and identification of the sources of sodium and chloride in natural waters of Illinois. Illinois State Geological Survey Open File Series 1, 15. Panno, S.V., Hackley, K.C., Hwang, H.H., Greenberg, S.E., Krapac, I.G., Landsberger, S., O'kelly, D.J., 2006. Characterization and identification of Na-Cl sources in ground water. Ground Water 44 (2), 176e187. n-Zapata, E., Ledesma-Ruiz, R., Harter, T., Ramírez, A.I., Mahlknecht, J., 2014. Paste Assessment of sources and fate of nitrate in shallow groundwater of an agricultural area by using a multi-tracer approach. Sci. Total Environ. 470, 855e864. Pichtel, J., Kuroiwa, K., Sawyerr, H.T., 2000. Distribution of Pb, Cd and Ba in soils and plants of two contaminated sites. Environ. Pollut. 110 (1), 171e178. Shi, G.L., Li, X., Feng, Y.C., Wang, Y.Q., Wu, J.H., Li, J., Zhu, T., 2009. Combined source apportionment, using positive matrix factorizationechemical mass balance and principal component analysis/multiple linear regressionechemical mass balance models. Atmos. Environ. 43 (18), 2929e2937. Shi, G.-L., Zeng, F., Li, X., Feng, Y.-C., Wang, Y.-Q., Liu, G.-X., Zhu, T., 2011. Estimated contributions and uncertainties of PCA/MLReCMB results: source apportionment for synthetic and ambient datasets. Atmos. Environ. 45 (17), 2811e2819. Smolders, E., Degryse, F., 2002. Fate and effect of zinc from tire debris in soil. Environ. Sci. Technol. 36 (17), 3706e3710. Song, Y., Dai, W., Shao, M., Liu, Y., Lu, S., Kuster, W., Goldan, P., 2008. Comparison of receptor models for source apportionment of volatile organic compounds in Beijing, China. Environ. Pollut. 156 (1), 174e183. Soundarrajan, C., Sivasankar, A., Maruthamuthu, S., Veluchamy, A., 2012. Improved lead recovery and sulphate removal from used lead acid battery through Electrokinetic technique. J. Hazard Mater. 217, 452e456. Thorpe, A., Harrison, R.M., 2008. Sources and properties of non-exhaust particulate matter from road traffic: a review. Sci. Total Environ. 400 (1e3), 270e282. Vile, M.A., Wieder, R.K., Nov ak, M., 2000. 200 years of Pb deposition throughout the Czech Republic: patterns and sources. Environ. Sci. Technol. 34 (1), 12e21. Vouitsis, E., Ntziachristos, L., Pistikopoulos, P., Samaras, Z., Chrysikou, L., Samara, C., Papadimitriou, C., Samaras, P., Sakellaropoulos, G., 2009. An investigation on the physical, chemical and ecotoxicological characteristics of particulate matter emitted from light-duty vehicles. Environ. Pollut. 157 (8e9), 2320e2327. Wang, Y.F., Huang, K.L., Li, C.T., Mi, H.H., Luo, J.H., Tsai, P.J., 2003. Emissions of fuel metals content from a diesel vehicle engine. Atmos. Environ. 37 (33), 4637e4643. Yatkin, S., Bayram, A., 2008. Source apportionment of PM10 and PM2. 5 using positive matrix factorization and chemical mass balance in Izmir, Turkey. Sci. Total Environ. 390 (1), 109e123. Zafra, C., Temprano, J., Tejero, I., 2016. The physical factors affecting heavy metals accumulated in the sediment deposited on road surfaces in dry weather: a review. Urban Water J. 14 (6), 639e649. Zafra, C., Temprano, J., Suarez, J., 2017. A simplified method for determining potential heavy metal loads washed-off by stormwater runoff from roaddeposited sediments. Sci. Total Environ. 601e602, 260e270. Zhang, W., Zhang, S., Wan, C., Yue, D., Ye, Y., Wang, X., 2008. Source diagnostics of polycyclic aromatic hydrocarbons in urban road runoff, dust, rain and canopy throughfall. Environ. Pollut. 153 (3), 594e601. Zhang, Y., Guo, C.S., Xu, J., Tian, Y.Z., Shi, G.L., Feng, Y.C., 2012. Potential source contributions and risk assessment of PAHs in sediments from Taihu Lake, China: comparison of three receptor models. Water Res. 46 (9), 3065e3073. Zhao, H.T., Li, X.Y., 2013. Understanding the relationship between heavy metals in road-deposited sediments and washoff particles in urban stormwater using simulated rainfall. J. Hazard Mater. 246, 267e276. Zhao, H.T., Li, X.Y., Wang, X.M., 2011. Heavy metal contents of road-deposited sediment along the urban-rural gradient around beijing and its potential contribution to runoff pollution. Environ. Sci. Technol. 45 (17), 7120e7127.