Motor Vehicle Wash-off Water as a Source of Phosphorus in ...

9

Original ArticleMotor Vehicle Wash-off Water as a Source of Phosphorus in

Roadway RunoffKeiko Wada a, Rie Simpson a, Naoyuki Kishimoto b, Naoko Takei c

a Institute of Water Environmental Research, Lake Biwa-Yodo River Water Quality Preservation Organization, Osaka, Japan

b Faculty of Science and Technology, Ryukoku University, Otsu, Japanc TORAY TECHNO CO., LTD., Otsu, Japan

ABSTRACTAlgal water blooms in lakes or reservoirs are often caused by an enrichment of phosphorus. Depend-ing on a bottom environment, dissolved phosphorus (phosphate) can be released into a water column from bottom sediments and accelerate algal and macrophyte nutrient dynamics. This study focuses on phosphorus pollutant loads in stormwater wash-off from roadways. Control of phosphorus discharge from non-point sources in urban areas is important for preventing water pollution. Sources of phos-phorus in pollutant loads were explored by comparing first flush runoff with road dust/water mixture and vehicle wash-off water, where characteristics of the particulate and dissolved portions, and the relationship between road dust and wash-off from vehicles were also discussed. It was clear that the concentration of dissolved phosphorus in the vehicle wash-off water was higher than that in the first flush runoff. One of the affecting factors of the dissolved phosphorus was inferred to be the nature of the additives in engine oils or certain types of engines.

Keywords: Roadway runoff, phosphorus, motor vehicle, engine oil additives, non-point source

INTRODUCTION

The Law Concerning Special Measures for the Preserva-tion of Lake Water Quality was enacted in 1984 to reduce water pollution in Japanese lakes and restore their water quality. The law regulates the environmental quality stan-dard values of pollutants such as organic matter, nitrogen, and phosphorus in lakes and reservoirs. Especially, instal-lation of sewerage systems and wastewater controls under the law had successfully decreased pollutant loads from point-sources, and had improved water quality of inflow rivers. However, the formulation of policies and regulations for non-point sources remains insufficient because of the dif-ficulties in tracking down the pollution sources.

In the case of agricultural non-point sources, agricultural policies related to land runoff have helped to minimize fertil-izer and pesticide uses on farms [1]. Roadway runoff includes

road dust, which contains pollutants from vehicles. These pollutants are composed of organic substances, suspended solids (SS), heavy metals, and other contaminants [2]. The first-flush runoff (FFR) from urban roadways is well known to have higher concentrations of pollutants than the following runoff [3–5]. Moreover, among non-point sources, roadway runoff is thought to be more influential in water pollution in urban areas because it directly flows into the rivers, lakes and enclosed water areas. The roadway runoff quality in-cludes not only harmful substances such as heavy metals and polycyclic aromatic hydrocarbons (PAHs), but also nutrient salts. Phosphorus, which is an essential mineral for aquatic plants and phytoplankton, often causes troublesome water blooms in phosphorus-limited lakes and reservoirs when its concentration reaches eutrophication levels [6]. Dissolved total phosphorus (D-TP) can be released into a water column from bottom sediments in anaerobic condition and acceler-

Corresponding author: Keiko Wada, E-mail: [email protected]: June 3, 2019, Accepted: September 30, 2019, Published online: February 10, 2020

Open Access

This is an open-access article distributed under the terms of the Creative Commons Attribution (CC BY) 4.0 License. http://

creativecommons.org/licenses/by/4.0/

Journal of Water and Environment Technology, Vol.18, No.1: 9–16, 2020doi: 10.2965/jwet.19-047

Journal of Water and Environment Technology, Vol. 18, No. 1, 2020 10

ates algae and macrophyte propagations. Accordingly, an occurrence of phytoplankton bloom implies phosphorus supplies from the bottom sediment and the surrounding watersheds to water bodies.

In the most of past studies on roadway runoffs, typical pol-lutants such as sediment, heavy metals, toxic hydrocarbons (ex. PAHs) [7–9] were main topics. Though total phosphorus (TP) was focused as nutrients in some cases [10], D-TP was hardly analyzed. Therefore, data of phosphorus in urban roadway runoff as a non-point source load are quite limited. It is important to understand the behavior of phosphorus including D-TP from urban roadways, for eutrophication control of receiving water bodies.

MATERIALS AND METHODS

Samples of first-flush runoff waterSamples of FFR from rainfall events on 5 roads in an urban

part of the Lake Biwa watershed were collected for 15 years (from 1999 to 2014). The area satisfied the conditions of rapid urbanization with increasing traffic volume. These general urban roadways had average daily traffic of approximately 15,000 to 36,000 vehicles and were covered with asphalt [11]. The locations of the survey stations are shown in Fig. 1. The detail specifications are given in the work of Wada et

al. [12]. The roadway runoff samples consisted of cumulative discharges of 2 to 7 mm starting from the onset of rain. All samples for analysis of TP were brought to the laboratory and divided by passing through a membrane filter with 0.45 µm pores (47 mm A045A047 ADVANTEC, Japan).

Samples of vehicle wash-off waterThe experiment was conducted on two types of unwashed

vehicles; diesel and gasoline motorcars. Vehicle wash-off water (VWW) was assumed to be included in the roadway runoff load and to flow down the surfaces of the vehicles to the road during rainfall events. The specifications of the vehicles used in the VWW experiment and the dates and conditions of the experiment are listed in Table 1. The amount of spray corresponding to the FFR from the surface area (37 to 39 m2) of the car body was calculated as a cumulative discharge of 5 mm (total volume 185 to 190 L) per car. The experiment was conducted with a simulated rainfall intensity of about 2.4 mm/h. First, 2 mm of distilled water was sprayed on the top and sides of the car body to simulate rain. Thereafter, 3 mm of distilled water was sprayed underneath the car to simulate the effects of water and mud splashed up by the wheels. All VWWs corresponding to the FFR were received by a clean polyethylene sheet lying under the car and were collected in a container (Fig. 2).

Fig. 1 Locations of survey stations for sampling first-flush runoff from urban roadways.

Journal of Water and Environment Technology, Vol. 18, No. 1, 2020 11

Samples of road dust adsorption testThe road dust adsorption test (RDA) was conducted using

the three VWW events (20140807, 20141031, and 20150507) samples and road dust. Road dust was collected from road-way gutter surfaces at St. 5 (see Fig. 1, 35.013°N 135.937°E) on 8 July 2015 and air-dried by room temperature. To prepare samples, the road dust was sieved twice, with 2 mm and then 0.2 mm screen meshes, to remove pebbles and coarse dusts. Each road dust sample was analyzed to measure the amount of phosphorus.

Then, VWW samples from 3 events (see Table 1) were mixed with the road dust at a solid to liquid ratio of 100 mg-dry weight/L (0.01 w/v%), which was equivalent to an av-

erage of SS concentration observed in the FFRs [12], and stirred with a stirrer (SM-600 Advantech) at 250 rpm for 1 hour. The particulate and dissolved fractions in the mixed samples were then separated using a membrane filter with 0.45 µm pores, and the D-TP in the filtered samples was quantified. Fig. 3 shows the flow of this procedure.

Chemical analysisPhosphorus concentrations in the sample solutions were

determined according to JIS K0102-46.3.1 [14] immediately after sample preparation. The amount of phosphorus in the road dust was measured according to Sediment Monitoring Methods II 4.9 [15].

Table 1 Specification of vehicle using experiment for vehicle wash-off water.Sampling date Previous rainfall amount a Dry weather period a Mileage (km)

(YYYYMMDD) (mm) (day) Type A b Type B c

20130415 16.5 9 248 9520131101 21 7 488 41320140415 9 11 81 440

20140807 d 4 11 1536 78120141003 4 8 148 493

20141031 d 13.5 4 11 6320150507 d 14 16 173 25620151014 22 12 145 251

a) Japan Meteorological Agency Database [13].b) Vehicle Type A: Toyota HiAce Van S-GL (2490-cc diesel engine), Engine oil: multi runner DH-2, Diesel particulate filter (DPF)-adaptive engine oil, 10W-30, (API CF-4 grade)c) Vehicle Type B: Toyota Regius Ace (2700-cc gasoline engine), Engine oil: motor multi SL/CF 10W-30 (API SL/CF grade)d) The conduct of the road dust adsorption test (RDA).

Fig. 2 Vehicle wash-off-water collections.

Journal of Water and Environment Technology, Vol. 18, No. 1, 2020 12

RESULTS AND DISCUSSION

Phosphorus concentrations in FFR and VWWFig. 4 shows the results for phosphorus concentrations in

the FFR and VWW. Concentrations of D-TP and particulate total phosphorus (P-TP) in the FFR were 0.026 ± 0.023 mg

P/L (average ± standard deviation) and 0.201 ± 0.149 mg P/L, respectively; the D-TP to P-TP ratio was 11:89. Clearly, the particulate portion reached approximately 90% of TP in the FFR. On the other hand, concentrations of D-TP and P-TP in the VWW were 0.066 ± 0.050 mg P/L and 0.074 ± 0.048 mg P/L, respectively. The ratio of D-TP to P-TP in the VWW was 47:53, i.e. the two concentrations were almost equal. In these results for FFR and VWW, phosphorus in the FFR was clearly dominated by particulate one. By contrast, dissolved phosphorus in the VWW occupied a higher percentage of TP than that in the FFR.

Usually, phosphorus is involved in soil mineral in road dust. Therefore, phosphorus in roadway runoff is mainly in particulate form. However, the D-TP concentration in the VWW was two to three times higher than that in the FFR and was nearly equal to the P-TP.

The ratios of D-TP to TP in FFR and VWW are shown in Fig. 5. The D-TP/TP ratios between FFR and VWW were significantly different (p < 0.05, t-test). In terms of emission processes, gasoline engine vehicles have adopted 3-way catalyst, and diesel engine vehicles have advanced engines with NOx reduction catalysts and diesel particulate filters (DPFs), which are very effective in purifying the emissions and trapping the finest soot particles produced as burning diesel’s fuel. However, quality standards for these engine oil is set to a minimum limit concentration of sulfate and phosphate compounds for preventing friction, rust, and oxidation in engines. These oils contain lubricants additives with phosphorus compounds (e.g. amine phosphate, acid phosphate, and zinc dialkyl dithiophosphate) as catalysts [16]. Therefore, it is possible that D-TP concentration in the

Fig. 3 Schematic diagram of road dust adsorption test.

Fig. 4 Phosphorus concentrations in first-flush runoff water (left; n = 30) and vehicle wash-off water (right; n = 16).

Journal of Water and Environment Technology, Vol. 18, No. 1, 2020 13

VWW is higher than that in the FFR.

Phosphorus pollution caused by different types of VWW

Fig. 6 shows the concentrations of P-TP and D-TP in VWW from vehicles powered by the two different types of fuel or engine oil. The average D-TP concentration of VWW was 0.066 mg P/L, namely 0.029 mg P/L for Type A (DPFs equipped diesel engine vehicle and API CF-4 grade engine oil) and 0.102 mg P/L for Type B (a gasoline engine vehicle and API SL/CF grade engine oil). The t-test showed a signifi-cant difference in the ratio of the D-TP to TP concentrations in the VWWs of both vehicles (p < 0.05).

Fig. 7 shows the relationship between D-TP loads calcu-lated per a vehicle and, previous rainfall amount and mile-ages of different vehicle types. The D-TP load from Type B tended to be higher than that from Type A. Especially, the two data of Type B (20130415 and 20131101) were much larger than others. The differences of the run condition of previous dry weather periods such as wind, dry deposits, and number of sloping path (e.g. hill start, run up downhill, etc.) might bring the relatively large variation. However, it was difficult to find the cause of these outlier. Thus the D-TP emission from the vehicles possibly depended on either the type of vehicle engine or additives in engine oils, though the run condition also influenced the D-TP loads.

Type B tended to show higher D-TP loads and lower P-TP loads than Type A as is shown in Fig. 6. As a result, the

TP load of Type B was slightly higher than that of Type A. In general, diesel vehicles emit larger amount of particulate matter than gasoline ones [17]. Accordingly, the higher P-TP load of Type A than that of Type B seemed to correspond with the emission rate of particulate matter. Cheung et al. reported lower emissions for the DPF-diesel and gasoline cars than for the diesel one without DPF, which highlighted the effectiveness of the DPF [18]. Presumably, TP loads of diesel vehicle without DPF would be suggested more higher than that of the DPF-diesel vehicle.

Based on the obtained results, the phosphorus loads of TP and D-TP from Type A were estimated to be 21.9 ± 8.7 and

Fig. 5 Comparison of the D-TP ratio in T-P concentration between FFR and VWW. The both samples results are significantly different ratios (p < 0.05; t-test).

Fig. 6 P-TP and D-TP concentrations in VWW of different vehicle types.

Journal of Water and Environment Technology, Vol. 18, No. 1, 2020 14

5.4 ± 3.2 mg P/vehicle, and those from Type B were estimated to be 30.7 ± 25.9 and 19.5 ± 17.3 mg P/vehicle, respectively. On average phosphorus loads of the TP, D-TP and P-TP in VWW from two types of vehicles were estimated to be 26.3 ± 19.8 and 12.4 ± 14.3 and 13.9 ± 9.0 mg P/vehicle, respec-tively. Based on these values, the phosphorus pollutant loads from vehicles in the Lake Biwa watershed were estimated to be 1.64 kg TP/d, 0.77 kg D-TP/d and 0.86 kg P-TP/d, re-spectively, by multiplying the average pollutant loads with the average daily traffic of Shiga Prefecture (62,221 cars/d as in 2010 [19]). This estimated load for TP was equivalent to 2.2% of the TP load from the urban area of Lake Biwa in 2015 (76 kg TP/d [20]).

Adsorption behavior of D-TP in vehicle wash-off waterVWW samples of 1 L from three events were mixed with

100 mg-dry weight road dust to elucidate the adsorption characteristics of D-TP in RDA from motor vehicles with diesel and gasoline engines. Table 2 shows the results of D-TP concentration in VWW and road dust mixed with VWWs from the two types of motor vehicles. The D-TP concentration eluted from the control sample (distilled water mixed road dust as 100 mg-dry weight/L) was 0.007 mg P/L.

Commonly, dissolved phosphorus in water is easily ad-

sorbed by soils and clay [21–23] or is tended to flocculate. However, the adsorption tests revealed that phosphorus was not adsorbed on road dust very much, namely the specific adsorption in the range from 0.03 to 0.05 mg/g. This suggests that the phosphate in engine oils and/or lubricant additives is less likely to be adsorbed by the fine soil on the road surface; our results clearly showed that phosphorus in the VWW remained in a dissolved form.

The average D-TP concentration in VWW and road dust mixture samples was 0.039 mg P/L. The D-TP concentration in VWW mixed road dust samples exceeded 0.03 mg P/L, which is the eutrophic level of T-P proposed by Sakamoto [24] and the OECD [25]. Phosphorus including D-TP from roadways was discussed in this research, for eutrophication control of receiving water bodies. This result was indicated that D-TP in VWW did not adsorbed on soils, but phospho-rus flowing into receiving water bodies could be sedimented on bottoms of lakes, reservoirs and/or closed water area. The availability of dissolved organic phosphorus compounds by phytoplankton has been intensively studied in water ecosystems. It was reported that microplankton promoted its phosphorus utilization by enhancing alkaline phosphatase activity under the condition of low phosphate concentrations [26], and the bacterial metabolism such as growth rate and

Fig. 7 Relationship between D-TP loads from vehicles and previous rain amount (left) or mileages (right).

Table 2 Adsorbed D-TP amount in vehicle wash-off water (VWW) from two type vehicles in VWW mixed road dust (ap-prox. 0.01 w/v%).

Type A Type B20140807 20141031 20150507 20140807 20141031 20150507

VWW (mg P/L) 0.043 0.018 0.014 0.040 0.052 0.051VWW+RD (mg P/L) 0.047 0.020 0.016 0.042 0.054 0.054Adsorbed phosphorus (mg P/g) 0.03 0.05 0.05 0.05 0.05 0.04

Journal of Water and Environment Technology, Vol. 18, No. 1, 2020 15

efficiency were positively related to total phosphorus concen-trations [27]. Depending on a bottom environment, D-TP can be released into a water column from bottom sediment and accelerate algal and macrophyte nutrient dynamics eventu-ally. From the perspective of anthropogenic eutrophication, we need to take preventive measures against direct discharge of pollutants from roadways into lakes and rivers nearby.

CONCLUSIONS

We examined the characteristics of phosphorus in roadway and vehicle runoff waters. The main results obtained in this study are as follows;

● Phosphorus in the FFR made up approximately 90% of the particulate portion. On the other hand, the D-TP concentration in the VWW was two to three times higher than that FFR and the average ratio of phos-phorus particles and dissolved phosphorus were about similar and is extremely solubility.

● The dissolved and particulate phosphorus concentrations in vehicle wash-off waters clearly differed between vehicles of different fuel types. One probable cause of dissolved phosphorus was suggested either the type of vehicle engine or additives in engine oils.

● The average phosphorus loads of the TP, D-TP and P-TP for the VWW were estimated to be 26.3 ± 19.8 and 12.4 ± 14.3 and 13.9 ± 9.0 mg P/vehicle, respectively. Moreover, the phosphorus pollutant loads from vehicle in the Lake Biwa watershed were calculated 1.64 kg TP/d, 0.77 kg D-TP/d and 0.86 kg P-TP/d, respectively.

Phosphorus substances are significant components caus-ing an eutrophication in water bodies. From our study, we found that phosphorus substances were not only compounds derived from soil and fertilizer but also is strongly dependent on motor vehicle. The results could be an effective factor as one of a parameter in phosphorus substances from road runoff. There is thus a need for increased research efforts that compare, for example, different size (compact and heavy, etc.) or maintenances status. Future investigations on the roadway runoff control of vehicles would be contribute to develop a new technology for automobile-related problem of pollution certainly.

ACKNOWLEDGEMENTS

This research was partially supported by Japanese Minis-try of Land, Infrastructure, Transport and Tourism.

REFERENCES

[1] Shiga Prefecture: Ordinance for Promotion of Environ-mentally Conscious Agriculture, 2003. https://www.pref.shiga.lg.jp/file/attachment/1009763.pdf [accessed on 23 December, 2019]

[2] Sartor JD, Boyd GB: Water pollution aspects of street surface contaminants, U.S. Environment Protection Agency Report, EPA-R2-72–081. U.S. EPA, Washing-ton DC, USA, 1972.

[3] Michael KS, Masoud K: First Flush Phenomenon Characterization, California Department of Transpor-tation Division of Environmental Analysis, University of California, CTSW-RT-05–073-02.6, LA, California, USA, 2005.

[4] Wada K, Yamanaka S, Yamamoto M, Toyooka K: The characteristics and measuring technique of refractory dissolved organic substances in urban runoff. Water Sci. Technol., 53(2), 193–201, 2006. PMID:16594338 doi:10.2166/wst.2006.053

[5] Wada K, Fujii S: Estimation of pollutant loads from urban roadway runoff. Water Sci. Technol., 61(2), 345–354, 2010. PMID:20107261 doi:10.2166/wst.2010.814

[6] Scheffer M: Chapter 2.3 Nutrient dynamics. In: Ecol-ogy of Shallow Lakes, Kluwer Academic Publishers, The Netherlands, Dordrecht/Boston/London, pp. 47–75, 2004.

[7] Asada M, Ohgaki S: Benzo(a)pyrene in urban runoff at the beginning of rainfall. J. Jpn. Soc. Water Environ., 19(11), 904–909, 1996. [in Japanese with English ab-stract] doi:10.2965/jswe.19.904

[8] Sakai A, Tanaka K, Sumiyama M: Water quality characteristics of urban area runoff and model analysis of non-point source pollutant runoff, J. Jpn. Sewage Works Association. Research journal, 34(413), 12–26, 1997. [in Japanese]

[9] United States Environmental Protection Agency: National Management Measures to Control Nonpoint Source Pollution from Urban Areas: United States En-vironmental Protection Agency Office of Water, Wash-ington, DC, USA, 20460 (4503F), EPA-841-B-05–004, 2005.

[10] Kayhanian M, Suverkropp C, Ruby A, Tsay K: Charac-terization and prediction of highway runoff constituent event mean concentration. J. Environ. Manage., 85(2), 279–295, 2007. PMID:17161904 doi:10.1016/j.jenv-man.2006.09.024

Journal of Water and Environment Technology, Vol. 18, No. 1, 2020 16

[11] Japanese Ministry of Land, Infrastructure, Transport and Tourism (Japanese MLIT): Roadways Traffic Census, 1999, 2005, 2010. http://www.mlit.go.jp/road/ir/ir-data/ir-data.html [accessed on 8 June, 2017, in Japanese]

[12] Wada K, Takei N, Sato T, Tsuno H: Sources of organic matter in first flush runoff from urban roadways. Water Sci. Technol., 72(7), 1234–1242, 2015. PMID:26398040 doi:10.2166/wst.2015.307

[13] Japan Meteorological Agency Database: Weather Statistics Information. http://www.data.jma.go.jp/obd/stats/etrn/index.php [accessed on 25 June, 2018, in Japanese]

[14] Japanese Industrial Standards Committee: Japanese Industrial Standard Testing Methods for Industrial Wastewater, JIS K0102. Japanese Standards Associa-tion, Tokyo, Japan, 2013.

[15] Japanese Ministry of the Environment (Japanese MOE): Sediment monitoring methods. 2012. http://www.env.go.jp/water/teishitsu- chousa/00_full.pdf [accessed on 4 July, 2015]

[16] JASO: Engine Oil Standards Implementation Panel, Operational manual for JASO M 355:2015, Japan Lu-bricating Oil Society, 65p, 2015. [in Japanese]

[17] Environmental Restoration and Conservation Agency: Information of Air Environment. https://www.erca.go.jp/yobou/taiki/taisaku/02_05.html [accessed on 28 July, 2019, in Japanese]

[18] Cheung KL, Polidori A, Ntziachristos L, Tzamkiozis T, Samaras Z, Cassee FR, Gerlofs M, Sioutas C: Chemi-cal characteristics and oxidative potential of particulate matter emissions from gasoline, diesel, and biodiesel cars. Environ. Sci. Technol., 43(16), 6334–6340, 2009. PMID:19746734 doi:10.1021/es900819t

[19] Tsuji K: Road traffic characteristics of Shiga Prefecture and future prospects. 35th Proceedings of Civil Engi-neers Workshop in Shiga Prefecture, 2014. https://www.pref.shiga.lg.jp/ippan/kendoseibi/gizyutsu/18824.html [accessed on 25 June, 2019, in Japanese]

[20] Shiga Prefecture: Chapter 3, Changes in loads inflow-ing to Lake Biwa (by the 7th Lake Biwa Water Quality Conservation Plan), Environment 2018 of Shiga, 2019. https://www.pref.shiga.lg.jp/file/attachment/4042980.pdf [accessed on 23 December, 2019]

[21] Arai S, Sugihara S: Behaviour of nitrogen and phos-phorus in the soil. Jpn. J. Water Pollution Research, 10(7), 402–406, 1987. [in Japanese]

[22] Matsumoto S: Soil general characteristics related to the purification capability. Jpn. J. Water Pollution Research, 10 (7), 388–393, 1987. [in Japanese]

[23] Wada K, Kishimoto N, Somiya I, Sato T, Tsuno H: Long-term evaluation of phosphorus removal purifica-tion performance and adsorption property of Akadama soil used in soil percolation treatment. J. Jpn. Soc. Water Environ., 37(2), 55–62, 2014. [in Japanese with English abstract] doi:10.2965/jswe.37.55

[24] Sakamoto M: Primary production by phytoplankton community in some Japanese lakes and its dependence on lake depth. Arch. Hydrobiol., 62, 1–28, 1966.

[25] OECD: Eutrophication of Water Monitoring, Assess-ment and Control, OECD, Paris, France, 1982.

[26] Chróst RJ, Overbeck J: Kinetics of alkaline phospha-tase activity and phosphorus availability for phyto-plankton and bacterioplankton in lake plu(North Ger-man Eutrophic Lake). Microb. Ecol., 13(3), 229–248, 1987. PMID:24213298 doi:10.1007/BF02025000

[27] Smith EM, Prairie YT: Bacterial metabolism and growth efficiency in lakes: The importance of phos-phorus availability. Limnol. Oceanogr., 49(1), 137–147, 2004. doi:10.4319/lo.2004.49.1.0137