Tropical climate effect on the toxic heavy metal pollutant course of
road-deposited sediments
*
Paula R.S. dos Santos
a,b, Glauber J.T. Fernandes
b, Edgar P. Moraes
c, Lucio F.F. Moreira
a,*aDepartment of Civil Engineering, Federal University of Rio Grande do Norte (UFRN), Natal, RN, 59078-970, Brazil bEnvironmental Chemistry Laboratory, CTGAS-ER, Natal, RN, 59063-400, Brazil
cChemometrics and Biological Chemistry Group (CBC), Institute of Chemistry, Federal University of Rio Grande do Norte (UFRN), Natal, RN, 59078-970,
Brazil
a r t i c l e i n f o
Article history: Received 12 March 2019 Received in revised form 25 April 2019
Accepted 9 May 2019 Available online 10 May 2019 Keywords:
Rainfall
Road-deposited sediment Toxic heavy metals Principal component analysis Natal
a b s t r a c t
In modern society, the intense vehicle traffic and the lack of effective mitigating strategies may adversely impact freshwater systems. Road-deposited sediments (RDS) accumulate a variety of toxic substances which are transported into nature during hydrologic events, mainly affecting water bodies through stormwater runoff. The aim of this study was to evaluate the RDS metal enrichment ratio between the end of wet season and the middle of the dry season for Cd, Cr, Cu, Fe, Mn, Ni, Pb and Zn in samples from Natal, Brazil. Twenty RDS, drainage system and river sediment samples were collected in the wet and dry seasons using a stainless-steel pan, brush and spatula. In the laboratory, the samples were submitted to acid digestion and heavy metal concentrations were measured by atomic absorption spectrometry (AAS). A consistent RDS enrichment by heavy metals in dry season samples was followed by an increase in the finest particle size fraction (D < 63mm). Maximum concentrations were 5, ND, 108, 23810, 83, ND, 77 and 150 mg kg-1for Cd, Cr, Cu, Fe, Mn, Ni, Pb and Zn, respectively. The RDS enrichment ratio was Cr(1.3 ), Cu(2.6 ), Fe(3.3 ), Mn(1.5 ), Pb(1.5 ) and Zn(2.1 ). The Geo-accumulation Index values showed that RDS were moderately polluted for Cu and slighted polluted for Zn and Pb. Principal Component Analysis (PCA) showed that the accumulation of toxic heavy metals decreased according to waterflow. © 2019 Elsevier Ltd. All rights reserved.
1. Introduction
Road-deposited sediments (RDS) are composed of particulates and contaminant residues derived from natural and anthropogenic sources, available for mobilization and transport to natural drainage systems during hydrologic events (Sutherland et al., 2012; Robertson and Taylor, 2007). RDS accumulate a wide variety of pollutants which adversely impact water, air and human health (Aryal et al., 2014; Loganathan et al., 2013; Mohammed et al., 2012). Given its ability to both associate with contaminants and be carried by surface runoff, RDS carry a high load of contaminants, especially toxic heavy metals and hazardous hydrocarbons (Sutherland et al., 2012; Murakami et al., 2008).
Vehicular traffic is considered the major RDS contamination origin, since it produces aerosol and particulates (Nguyen, 2016; Li
et al., 2015) sourced by vehicles (tires, lubricants, oils, brakes, exhaust) and road surface wear (Sutherland et al., 2012; Taylor and Owens, 2009). Studies demonstrate that these substances are the major toxic heavy metal sources in urban areas (Hjortenkrans et al., 2008; Charlesworth et al., 2003). Most of the RDS metals contam-ination studies have investigated Cd, Cr, Cu, Ni, Pb and Zn contamination levels (Zhang et al., 2015a; Li et al., 2015).
Several studies associate the RDS metal enrichment to weather factors like rainfall and wind. For example, Robertson & Taylor (2007) observed that metal concentrations in RDS were highly influenced by rainfall. These observations may be explained by the sorption ability of metals like Pb, Zn and Cu to thefinest particle size fraction. As rainfall occurs, surface runoff transports thefine sediment to the water bodies (Fergusson and Kim, 1991).
In Brazil,McAlister et al. (2005)investigated the levels of Fe, Mn, Zn, Cu, Pb, Cr and Ni in RDS from areas with high traffic density in Rio de Janeiro state. The study detected oxalate in Niteroi city, a harmful substance associated to thefine fraction (<2
m
m).Poleto et al. (2009)observed high levels of Cr, Cu, Ni, Pb and Zn in RDS samples collected from 20 municipalities in the Southern region of*This paper has been recommended for acceptance by Dr. Yong Sik Ok.
* Corresponding author.
E-mail address:lucio@ct.ufrn.br(L.F.F. Moreira).
Contents lists available atScienceDirect
Environmental Pollution
j o u r n a l h o m e p a g e :w w w . e l s e v i e r . c o m/ l o ca t e / e n v p o l
https://doi.org/10.1016/j.envpol.2019.05.043
Brazil. In these studies, however, the influence of weather factors on the RDS enrichment by metals was not investigated. The weather seasonal effect on RDS enrichment by metals is a gap that was explored in this study.
A recent monitoring study aimed at analyzing bed sediment contamination by metals in the Pitimbu River detected high levels of Pb and Fe. This raised concern about the delivery of RDS and associated contaminants to the river. The Pitimbu River discharge ranged between 0.63 (dry season) and 10.6 m3/s (storm event) from the BR-101 in 2011 (Oliveira, 2012).
In recent decades, the increasing number of circulating vehicles in Brazilian medium-sized cities has raised concern because of the aquatic systems’ vulnerability to contamination, since most of them are used for water supply (Duggan and Williams, 1977; Robertson et al., 2003). Because information about RDS contamination is crucial for management strategy planning, this study aims to pro-vide data that could be useful for government administration (Billota and Brazier, 2008; Charlesworth and Lees, 1999; Charlesworth et al., 2003).
In this context, the work reported herein investigated the in-fluence of weather factors on RDS enrichment by Cd, Cr, Cu, Fe, Mn, Ni, Pb and Zn for two highways in the urban area of Natal, Brazil, including RDS, drainage system and river sediment samples. 2. Materials and methods
2.1. Study location
The urban area of Natal is located at latitude 54704200S and longitude 351203200W; it is the capital of Rio Grande do Norte state, located in the east corner of the South American Continent (Fig. 1) and lies within the Pitimbu River drainage basin (approxi-mately 126 km2), an important water supply source for approxi-mately 300,000 people of Natal. The Pitimbu River basin has been subjected to adverse impacts linked to an expanding urbanization process. The urban area of Natal (167 km2) is the most populated in the state with 885,180 inhabitants (IBGE, 2017). The climate in the region is classified as humid tropical with a temperature range between 23 and 32C and average annual rainfall of 1642 mm. High-intensity rainfall is likely to occur in the region during the months of MayeJuly due to the seasonal regime. Since the 1990s,
urban development and population growth in the region have induced the construction and enlargement of highways. The official number of vehicles that currently circulate in Natal is 225,994, but the growth rate of more than 100% in the last 10 years indicate a rapid increase in vehicular traffic. Consequently, an increasing load of traffic contaminants are likely to accumulate on the road sur-faces, being available to be transported to the receiving water bodies during hydrologic events.
2.2. Sampling procedures and chemical analysis
The study area contains two major highways in the urban area of Natal where sampling sites were randomly selected: the BR-101 (high density residential-commercial), and BR-304 (mixed low density industrial and natural vegetation). Each selected sampling site was likely to possess some accumulation of RDS. Given that RDS removal by street-sweeping during the study period was lacking, it may be assumed that RDS contamination in BR-304 was related to vehicular traffic. On the other hand, a variety of waste material (paper, plastic, wood, metal) mixed to the sediment in the BR-101 roadway gutter may indicate the influence of other sources in the RDS contamination.
Twenty samples (RDS, n¼ 8; drainage system, n ¼ 6; river sediment, n¼ 6) were collected in two sampling campaigns. The first campaign was conducted in August-2017 (end of the rainy season) and the second in December-2017 (dry season). The RDS sampling area was 1 m 1 m (along the roadway gutter). RDS collection was done by sweeping with a stainless-steel pan and polyethylene brush, similar to previous studies (Wang et al., 2011; Nguyen, 2016). This enables a comparison of results with other studies. Sediment samples were taken in the drainage system adjacent to the highway by using an acid-cleaned plastic spatula. The pan and brush were cleaned with distilled water and dried after each sample collection to avoid cross contamination. Bed sediment samples from the Pitimbu river were collected using a Van Veen grab. Samples were placed in propylene airtight zip bags and stored.
In the laboratory, samples were air-dried and mixed by using agate mortars and pestles. The pebbles and organic matter were removed by using a 0.45 mm sieve. After sieving to obtain 5 g of the finest fraction, samples were dried in a hot air oven for 24 h at
100C. Nitric acid (Labsynth, SP, Brazil) was used to digest 5 g of sample in a microwave following theEPA 3051AMethod for 30 min at 190C. Digested solutions were placed in 50 ml vessels. Lastly, Fe, Mn, Zn, Cu, Pb, Cr and Ni concentrations (AccuStandard, CT, USA) were analyzed in Flame Atomic Absorption Spectrometry (FAAS) with an air-acetyleneflame (Shimadzu, Tokyo, Japan).
2.3. Assessing metal contamination 2.3.1. Geo-accumulation index
Heavy metal pollution in sediments can be assessed by using the concept of Geo-accumulation Index (Igeo), originally proposed by
Muller (1979) to quantitatively assess the metal pollution by comparing the baseline metal concentration with a reference background value. This method minimizes the effect of particle size on the geochemical variability as it considers the use of a single reference value. It is calculated by the following equation: Igeo¼ log2
Cn
1:5 Bn
where Cnis the baseline metal concentration in the RDS sample and
Bnis the geochemical background value. The constant 1.5 reduces
the baseline values by a factor of about 33%, which reduces the effect of natural variations in the metal values. It increases the model capacity to assess the anthropogenic influence. In this study, the calculation of Igeo considered the mean concentrations in the background sites measured in a previous study in the watershed (Moreira et al., 2014). This index is used to classify the level of pollution in seven categories, which are described as follows: class G0, unpolluted (Igeo 0); class G1, unpolluted to moderately polluted (0< Igeo 1); class G2, moderately polluted (1< Igeo 2); class G3, moderately to strongly polluted (2< Igeo 3); class G4, strongly polluted (3 < Igeo 4); class G5, strongly to very strongly polluted (4< Igeo 5); and class G6, very strongly polluted (5< Igeo). The reference values in soil for Igeo and EF indexes calculation were obtained in previous studies. Metal concentrations in soils are linked to geochemical and genesis pro-cesses. The reference values used in this study correspond to the Yellow Latosol predominance in the region (in mg kg-1): Cd¼ 0.8; Cr¼ 48; Cu ¼ 19; Fe ¼ 43,500; Mn ¼ 24,000; Ni ¼ 18; Pb ¼ 25 and Zn¼ 44.
2.3.2. Pollution load index (PLI)
The pollution load index has been originally used in monitoring programs as an index of bioavailability of contaminants for or-ganisms in coastal waters (Tomlinson et al., 1980). It standardizes the data by applying the values obtained by dividing each metal concentration by a reference value of each contaminant. It is calculated as follows:
PLI¼ nffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiðCF1x CF2x… CFnÞ
p
where. CF ¼ Cmetal=Creference
CFiis the contamination factor for metal i; Cmetalis the metal
concentration in the sample; Creference is the reference value of
metal. PLI reflects the integrated pollution of the metals which compose the sample. Pollution intensity is classified in classes on the basis of the PLI values as follows: non-pollution, PLI values 1; slight pollution, 1< PLI 2; moderate pollution, 2 < PLI 3; and heavy pollution, PLI>3.
2.4. Multivariate analysis
Data import, pre-treatment and multivariate procedures were
carried out using R software (R Core Team, 2018) with the PLS package version 2.6e0 (Mevik, Wehrens and Liland, 2016) and the ChemometricsWithR package (Wehrens, 2011). All data processing was performed using the free R 3.5.1 software from the R Foun-dation for Statistical Computing (2018). Raw metal composition data were pre-processed by autoscaling. Exploratory analysis was performed using Principal Component Analisys (PCA), in which raw data are transformed into new factors that are adjusted capturing the greatest variance in the data set (Hibbert, 2016).
3. Results and discussion 3.1. Particle size composition
The 60 days of antecedent rainfall which occurred prior to the first sampling campaign influenced the RDS particle size distribu-tion, because most of thefinest RDS sediment (D < 63
m
m) was transported during hydrologic events. Significant differences were observed in the RDS particle-size distribution between the two sampling campaigns. For the first campaign (wet season), the particle-size fraction< 63m
m proportion of the total sediment mass was small (5% on average). This coarsening was possibly caused by the surface runoff (Robertson and Taylor, 2007; Fergusson and Kim, 1991; Hamilton et al., 1984). For this reason, 177m
m (fine sand) particle-size fraction was used to determine metal concentrations for all samples. Conversely, the RDS particle-size fraction<63m
m (siltþ clay) proportion was high (23% on average) for the second campaign (dry season). This means that a consistent increase in thefinest RDS fraction occurred, possibly linked to RDS accumulation during the study period. Becausefiner RDS particle-size fractions have more ability to absorb heavy metals, RDS enrichment was highly influenced by the increase in these fractions (Robertson and Taylor, 2007).Traffic density monitoring data available for the 101 and BR-304 in 2017 indicated that the average traffic densities were 80,000 and 6269 vehicles/day (working days), respectively. The BR-304 traffic is composed of automobiles (72%), buses (16%) and trucks (12%), while the BR-101 traffic is mostly composed by automobiles (90%). Given the traffic characteristics and the absence of an effective pavement cleaning management strategy, it is possible that RDS contamination has a significant impact on river and water quality.
Daily precipitation for 2017 is presented inFig. 2(UFRN Weather Station). It demonstrates that the antecedent 60 days of rainfall prior to the sampling campaigns were 570 mm and 51.3 mm, respectively. The precipitation amount prior to thefirst campaign suggests that high intensity rainfall was likely to occur in July-eAugust. Conversely, the low amount of precipitation prior to the second campaign shows that high intensity events were lacking during the period of AugusteDecember. It may be assumed that the absence of high-intensity rainfall events during this period was the major influence for the RDS enrichment by heavy metals.
3.2. Metal concentrations
Metal concentrations for the two sampling campaigns are pre-sented inTable 1. A consistent enrichment of the RDS by metals was observed during the study period for each sampling site. The highest Cu concentration was observed in the dry season RDS sample 2 (108.4 mg kg-1). Fe concentrations in RDS varied between 7224.8 (wet season RDS 1 - WRDS1) and 23,810.4 (dry season RDS 1 - DRDS1). The highest metal levels for Pb, Zn, Mn and Cu were observed in the second campaign, possibly due to the enrichment process during the period between campaigns. A consistent tem-poral RDS enrichment by Mn was observed for all sites. The highest
Zn concentration was seen for the dry season RDS sample 2 (149.8 mg kg-1). Pb concentrations ranged between 49.7 and 119.4 mg kg-1 (WRDS1 and dry season drainage system 1 DDS1, respectively), which reflects the RDS enrichment from vehicular sources during the low rainfall period.Massadeh and Snook (2002)
observed that Pb and Zn concentrations in the RDS were highly associated with traffic conditions. Cr was only observed during that time, with the highest concentration in WRDS1 (33.7 mg kg-1). The highest metal concentrations were observed in DRDS2 for Cu, Mn and Zn. Previous studies indicate that heavy metals in RDS are mainly sourced by vehicles (Charlesworth et al., 2003; Davis and Birch, 2011; Loganathan et al., 2013; Faiz et al., 2009). Pb is sourced from vehicle exhaustions (Robertson et al., 2003; Zhu et al., 2001), and Zn is sourced from tire residues and brake linings. As shown inTable 1, the enrichment ratio for Cu, Mn, Zn, Fe, Pb and Cr
were 2.6 ; 1.5 ; 2.1 ; 3.3 ; 1.5 and 1.3 , respectively, and was mainly sourced by vehicular traffic. It shows that the enrich-ment was highest for Fe, followed by Cu, Zn, Mn and Pb, and lowest for Cr. The values of geo-accumulated index Igeo are presented in
Table 2for Campaigns 1 and 2 (C1 and C2). The highest Igeo level was seen for Cu in C2 and indicates moderate contamination. Furthermore, it was seen that Zn and Pb had a low contamination level in the RDS. RDS samples were not contaminated by Cd, Mn, Fe and Cr. The average pollution load indexes are presented inTable 1
for each metal. The results indicated moderate pollution for Cu, and slight pollution for Zn and Pb. According to the PLI assessment, Cd, Mn, Fe and Cr were in the non-pollution category. Despite the limited number of samples, a consistent reduction in the level of contaminants was observed for toxic heavy metal in the Drainage System (DS) and River Sediment (RS) samples. This is probably due Fig. 2. Daily precipitation in 2017 for the study area (UFRN Weather Station data). Red arrows indicate the dates of the sampling campaigns in the study. (For interpretation of the references to colour in thisfigure legend, the reader is referred to the web version of this article.)
Table 1
Heavy-metal concentrations (mg kg-1) in the wet season (W) and dry season (D) for Road-Deposited Sediment (RDS), Drainage System (DS) and River Sediment (RS).<LQ
indicates a lower metal concentration than the quantification capacity.
Sample Cd Cu Mn Zn Fe Pb Cr WRDS1 <LQ <LQ 13.6 <LQ 7224.8 49.7 <LQ WRDS2 4.8 41.7 55.2 69.8 14680.3 52.3 <LQ WRDS3 <LQ 27.9 49.8 37.9 11657.4 <LQ <LQ WRDS4 <LQ 71.9 86.3 129.4 16408.4 58.4 19.5 WDS1 <LQ <LQ 10.4 <LQ 2637.3 <LQ <LQ WDS2 5.1 <LQ <LQ <LQ 3406.8 <LQ <LQ WDS3 3.2 <LQ <LQ <LQ 2804.6 <LQ <LQ WRS1 <LQ <LQ <LQ <LQ 3284.1 <LQ <LQ WRS2 <LQ <LQ 9.9 <LQ 4526.9 <LQ <LQ WRS3 <LQ <LQ 3.1 <LQ 2431.4 <LQ <LQ DRDS1 <LQ 26.9 40.5 31.8 23810.4 56.8 33.7 DRDS2 <LQ 108.4 82.8 149.8 13630.9 60.1 27.6 DRDS3 <LQ 104.5 70.9 75.9 18814.8 76.9 24.7 DRDS4 <LQ 106.4 74.6 89.8 21845.3 74.8 24.6 DDS1 <LQ 59.7 49.7 95.5 12091.1 119.4 21.9 DDS2 <LQ 64.9 51.6 40.3 10450.0 74.3 23.6 DDS3 <LQ 53.6 48.4 74.0 10258.3 55.1 19.4 DRS1 <LQ <LQ 15.7 <LQ 2679.5 <LQ <LQ DRS2 <LQ <LQ 3.4 <LQ 2156.8 <LQ <LQ DRS3 <LQ <LQ 4.9 <LQ 2845.6 <LQ <LQ Igeo (C1) G0 G1 G0 G1 G0 G1 G0 Igeo (C2) G0 G2 G0 G1 G0 G1 G0
to the washing effect, because contaminants were flushed and transported by theflow to the river during rainfall events.
Table 2shows the RDS contamination levels obtained in other studies (Li et al., 2015; Zhang et al., 2015; Nguyen, 2016). Cr was only detected in one sample for thefirst campaign. Nevertheless, mea-surement data for the second campaign indicate consistent RDS contamination by Cr, with an average concentration of 29 mg kg-1. Cadmium was only detected in three samples from the first campaign, possibly originating from heavy truck traffic oil leakage. Ni was not detected in the RDS. The observations indicated that RDS metal enrichment was mainly due to vehicular traffic during the study period. RDS enrichment is a continuous temporal process.
3.3. Temporal variability
There was a clear tendency for the enrichment of Cu, Mn, Fe, Pb and Cr during the dry season period (Fig. 3). Our results showed that a clear separation due to tropical climate effect in the wet season (green) and dry season (red) was established during thefirst component PC 1 [72.8%], while the second component PC 2 [14.1] provided the clustering of samples due to the Cd presence condi-tion. In other words, the principal components were calculated by capturing the greatest variance in the data base and could classify the samples as a function of the season. The absence of Cr in RDS for the wet season campaign was possibly due to the metal's ability to mobilize. It is known that the Cr, Cu and Pb sorption processes occur in the presence of humified organic matter, Al oxides and, in
some cases, depends on cation-exchange capacity of the soil (Costa et al., 2017). These sediments' binding characteristics are likely to be important factors controlling the enrichment process. The enrichment by Cu and Zn confirms that these metals are sourced from vehicular sources such as bodywork residues, tires and brake pads. Low concentrations of Pb in the RDS for thefirst campaign confirm the study ofSutherland et al. (2012)for the case of insig-nificant fine particle-size sediment proportion.
Although the observations showed a significant temporal vari-ability in RDS grain-size characteristics, it is reasonable to assume that little variation occurred in the traffic characteristics during the study period. Observations showed that accumulation of Pb, Zn, Cr, Mn and Cu in the RDS originated from vehicle exhaust and wear of mobile parts. The BR-304 highway is used for transportation of a big variety of manufacturing products from Southeastern Brazil. The 63
m
m particle size in the RDS washed out during hydrologic events degrades the Pitimbu Riverfluvial system because of its high metal sorption ability. The introduction of these contaminants (in particulate bound or dissolved phase) to the river adversely im-pacts water resources and aquatic biota. Studies have shown that fine sediment is detrimental to potable water because of the contamination by heavy metals (Sutherland et al., 2012; Robertson and Taylor, 2007).3.4. Toxic heavy metal pollutant course
To obtain more detailed information about the pollution course, Table 2
RDS concentrations results in several studies.
City/Reference Cu Cd Pb Cr Ni Mn Zn Fe Particle size (mm)/Samples Goiania, Brazil/Silva,
2014
57 4.5 62.5 ND NA 148 49.3 13019 63e250/36 Dresden, Germany,Zhang et al., 2015 210 0.16 NA NA NA NA 500 NA 100e400/6 Manchester, UK/Taylor and Owens, 2009. 89 NA 92 NA NA 135.4 165 56983 63e300/9 Sydney, Australia/Nguyen, 2016. 266 0.2 165 42 14 567 544 19645 <2000/11 Beijing Olympic Park, China/Li et al. (2015) 126 3.1 510.7 NA NA NA 51.4 NA <2000/8 Punjab, Pakistan/Faiz et al. (2009). 52 5.0 104 NA 23 NA 116 NA <125/13 Rio Grande do Sul, Brazil/Poleto et al. (2009) 114 NA 52 157 62 NA 256 NA <63/20 Rio de Janeiro, Brazil/McAllister et al., 2007 290 NA 200e700 202 73.8 500e1200 2612 69000 <63/18 Hangzhou, China/Zhang et al. (2015). 177 3.0 198.1 147.5 47.4 712.3 1254 NA <75/35 The present study,
Natal-Brazil
108.4 5.1 119.4 33.7 ND 86.3 149.8 23810.4 <177/07
Fig. 3. Multivariate data analysis comparing wet and dry season samples. Principal Component Analysis (PCA) of variables in Road-deposited Sediment and Drainage System samples with autoscaling.
groups were compared by season:
RDS vs Drainage System vs River Sediment in order to observe the toxic heavy metalflow (Figs. 4 and 5). Relationships between metal elements in the PCA analysis can indicate possible originating sources and relationships between metals. Metal loading vectors suggest a decrease in contamination in the RDS, drainage system and river sediment samples. This means that probably RDS enrichment by metals was the main pollution source driven to the freshwater system. Cadmium correlation with the other metals was negative and indicates that Cd contamination was site-specific and associated with vehicles sources, possibly lubricating oil leakage.
Table 2 presents metal concentrations for RDS in Brazil and other countries. It can be seen that the concentrations of all metals in Natal are considered to present a low contamination level when compared to other locations. A significant influence of the RDS particle size can also be seen in the results.
4. Conclusions
This study focused on the contamination characteristics in RDS for wet and dry weather conditions and vehicular traffic. The re-sults showed that RDS contamination is a continuous and cumu-lative process, which is mostly influenced by vehicular traffic density. It was observed that RDS enrichment by heavy metals was followed by an increase in thefinest D < 63
m
m particle size in the RDS composition. RDS contamination by Zn, Cu, Mn, Cr and Pb was mainly sourced from vehicular traffic, namely from the wear ofmoving parts, oils, brake lines and vehicle exhaust. The data re-ported herein highlight the importance of establishing RDS man-agement strategies in line with rainfall seasonality. The RDS enrichment ratios obtained in this study [Cr(1.3 ), Cu(2.6 ), Fe(3.3 ), Mn(1.5 ), Pb(1.5 ) and Zn(2.1 )] demonstrated that contamination is an ongoing process. Finally, because river sedi-ment contamination is cumulative over time, effective manage-ment strategies regarding pollution control (traffic controlling measures, road sweeping and maintenance, public awareness, etc.) are of crucial importance to protect the aquatic system from degradation.
Acknowledgements
The authors acknowledge the support of the CTGAS-ER for the laboratory analysis.
References
Aryal, R.K., Duong, T.T.T., Lee, B., Hossain, M.B., Kandel, D., Kamruzzaman, M., Beecham, S., Chong, M.N., 2014. Organic matter composition variability in road sediment and its role in binding heavy metals. Sustain. Environ. Res. 24, 81e91. Billota, G.S., Brazier, R.E., 2008. Understanding the influence of suspended solids on water quality and aquatic biota. Water Res. 42, 2849e2861.https://doi.org/ 10.1016/j.watres.2008.03.018 2008.
Charlesworth, S.M., Everett, M., Mc Carthy, R.A., 2003. Comparative study of heavy metal distribution in deposited street dusts in a large and small urban area: birmingham and Coventry, UK. Environ. Int. 29, 563e573.
Charlesworth, S.M., Lees, J.A., 1999. The distribution of heavy metals in deposited urban dusts and sediments, Coventry, England. Environ. Geochem. Health 21,
Fig. 4. PCA results of PC1 score plot versus PC2, and loading plot of metal variables in Road-deposited Sediment, Drainage System and River Sediment during the wet season.
97e115.
Costa, J.J.G., Reigosa, M.J., Matias, J.M., Covelo, E.F., 2017. Soil Cd, Cr, Cu, Ni, Pb and Zn sorption and retention models using SVM: variable selection and competitive model. Sci. Total Environ. 593, 508e522.
Davis, B.S., Birch, G.S., 2011. Distribution of bulk atmospheric deposition of heavy metals in metropolitan sydney, Australia. Water Air Soil Pollut. 214, 147e162.
https://doi.org/10.1007/s11270-010-0411-3.
Duggan, M.J., Williams, S., 1977. Lead-in-dust in city streets. Sci. Total Environ. 7, 91e97.
Faiz, Y., Tufail, M., Javed, T., Chaudhry, M.M., 2009. Road dust pollution of Cd, Cu, Ni, Pb and Zn along islamabad expressway, Pakistan. Microchem. J. 92, 186e192.
https://doi.org/10.1016/j.microc.2009.03.009.
Fergusson, J.E., Kim, N.D., 1991. Trace elements in street and house dusts: sources and speciation. Sci. Total Environ. 100, 125e150. https://doi.org/10.1016/0048-9697(91)90376-P.
Hamilton, R.S., Revitt, D.M., Warren, R.S., 1984. Levels and physico-chemical asso-ciation of Cd, Cu, Pb and Zn in road sediments. Sci. Total Environ. 33, 59e74.
https://doi.org/10.1016/0048-9697(84)90381-4.
Hibbert, D.B., 2016. Vocabulary of concepts and terms in chemometrics (IUPAC Recommendations 2016). Pure Appl. Chem. 88 (4), 407e443.https://doi.org/10. 1515/pac-2015-0605.
Hjortenkrans, D.S.T., Bergback, B.G., Haggerud, A.V., 2008. Transversal emission patterns and leachability of heavy metals in road side soils. J. Environ. Monit. 10, 739e746.https://doi.org/10.1039/b804634d.
IBGE Instituto Brasileiro de Geografia e Estatística, 2017. Dados do censo. Available at:www.ibge.gov.br.
Li, H., Shi, A., Zhang, X., 2015. Particle size distribution and characteristics of heavy metals in road-deposited sediments from Beijing Olympic Park. J. Environ. Sci. 32, 228e237.https://doi.org/10.1016/j.jes.2014.11.014.
Loganathan, P., Vigneswaran, S., Kandasamy, J., 2013. Road-deposited sediment pollutants: a critical review of their characteristics, source apportionment, and management. Crit. Rev. Environ. Sci. Technol. 43, 1315e1348.https://doi.org/10. 1080/10643389.2011.644222.
Massadeh, A.M., Snook, R.D., 2002. Determination of Pb and Cd in road dusts over the period in which Pb was removed from in the UK. J. Environ. Monit. 4, 567e572.https://doi.org/10.1039/B203905M.
Mc Alister, J.J., Smith, B.J., Neto, J.B., Simpson, J.K., 2005. Geochemical distribution and bioavailability of heavy metals and oxalate in street sediments from Rio de Janeiro, Brazil: a preliminary investigation. Environ. Geochem. Health 27, 429e441.https://doi.org/10.1007/s10653-005-2672-0.
Mevik, B.H., Wehrens, R., Liland, K.H., 2016. Pls: partial least squares and principal component regression. R package version 2.6-0. https://CRAN.R-project.org/ package¼pls.
Mohammed, T., Loganathan, P., Kinsela, A., Vigneswaran, S., Kandasamy, J., 2012. Enrichment, inter-relationship and fractionation of heavy metals in road-deposited sediments of Sydney, Australia. Soil Res. 50, 229e238.https://doi. org/10.1071/sr12010.
Moreira, L.F.F., Adamowski, J., Gaskin, S., Saraiva, A., 2014. The Impact of Poor Governance on Water and Sediment Quality: A Case Study in the Pitimbu River, Brazil. Proceedings of ICWRS2014.https://doi.org/10.5194/piahs-364-429-2014
(IAHS Publications 364, 2014).
Muller, G., 1979. Schwermetalle in den Sedimenten des Rheins: Veranderungen seit 1971. Umschau 79, 778e783.
Murakami, M., Sato, N., Anegawa, A., 2008. Multiple evaluations of the removal of pollutants in road runoff by soil infiltration. Water Res. 42, 2745e2755.https:// doi.org/10.1016/j.watres.2008.02.004.
Nguyen, T.C., 2016. Pollutants in Road-Deposited Sediments: Characteristics, Mobility, Bioavailability and Remediation. Thesis. School of Civil and Environ-mental Engineering, Faculty of Engineering& Information Technology, Uni-versity of Technology Sydney., New South Wales, Australia.
Oliveira, A.S., 2012. Uso e ocupaç~ao do solo e a concentraç~ao de metais pesados no sedimento e naagua: bacia do rio Pitimbu. Dissertaç~ao, Programa de Pos-graduaç~ao em Engenharia Sanitaria. UFRN.
Poleto, C., Bortoluzzi, E.C., Charlesworth, S.M., Merten, G.H., 2009. Urban sediment particle size and pollutants in Southern Brazil. J. Soils Sediments 9, 317.https:// doi.org/10.1007/s11368-009-0102-0.
R Core Team, 2018. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria.https://www.R-project. org/.
Robertson, D.J., Taylor, K.G., 2007. Temporal variability of metal contamination in urban road-deposited sediment in manchester, UK: implications for urban pollution monitoring. Water Air Soil Pollut. 186, 209e220.https://doi.org/10. 1007/s11270-007-9478-x.
Robertson, D.J., Taylor, K.G., Hoon, S.R., 2003. Geochemical and mineral character-ization of urban sediment particles, Manchester, UK. Appl. Geochem. 18, 269e282.https://doi.org/10.1016/S0883-2927(02)00125-7.
Sutherland, R.A., Tack, F.M., Ziegler, A.D., 2012. Road-deposited sediments in an urban environment: afirst look at sequentially extracted element loads in grain size fractions. J. Hazard Mater. 225, 54e62.https://doi.org/10.1016/j.jhazmat. 2012.04.066.
Taylor, K.G., Owens, P.N., 2009. Sediments in urban river basins: a review of sedimentecontaminant dynamics in an environmental system conditioned by human activities. J. Soils Sediments 9, 281e303. https://doi.org/10.1007/s11368-009-0103-z.
Tomlinson, D.L., Wilson, J.G., Harris, C.R., Jeffrey, D.W., 1980. Problems in the assessment of heavy-metal levels in estuaries and the formation of a pollution index. Hel- gol€an der meeresun tersuchun gen 33 (1), 566.https://doi.org/10. 1007/BF02414780.
United States Environmental Protection Agency, EPA 3051A, 2007. Microwave assisted acid digestion of sediments, sludges, soils, and oils. Available at:www. epa.gov/hw-sw846/sw-846-test-method-3051a-microwave-assisted-acid-digestion-sediments-sludges-soils-and-oils.
Wang, W., Huang, M., Kang, Y., 2011. Polycyclic aromatic hydrocarbons (PAHs) in urban surface dust of Guangzhou, China: status, sources and human health risk assessment. Sci. Total Environ. 409, 4519e4527. https://doi.org/10.1016/j. scitotenv.2011.07.030.
Wehrens, R., 2011. Chemometrics with R: Multivariate Data Analysis in the Natural Sciences and Life Sciences. Springer, Heidelberg. https://doi.org/10.1007/978-3-642-17841-2.
Zhang, J., Hua, P., Krebs, P., 2015. The build-up dynamic and chemical fractionation of Cu, Zn and Cd in road-deposited sediment. Sci. Total Environ. 532, 723e732.
https://doi.org/10.1016/j.scitotenv.2015.06.074.
Zhu, B.Q., Chen, Y.W., Peng, J.H., 2001. Lead isotope geochemistry of the urban environment in the Pearl river delta. Appl. Geochem. 16, 409e417.https://doi. org/10.1016/S0883-2927(00)00047-0.
