大气污染防治

Road fugitive dust emission character-istics in Beijing during OlympicsGame 2008 in Beijing，China

Fan Shou-bin1，2，Tian Gang1，Li Gang1，Huang Yu-hu1，Qin Jian-ping1，Cheng Shui-yuan2

1. Beijing Municipal Research Academy of Environmental Protection，Beijing 100037，China.

2. College of Environmental and Energy Engineering，Beijing University of Technology，Beijing 100124，China.

Abstract：Eighty road dust-fall（DF）monitoring sites and 14 background monitoring sites were established in the Beijing metropolitan area，and monitoring was conducted from January 2006 to December 2008.The dust-fall attributable to roads（ΔDF）showed a clear decline from 2006 to 2008.Dust-fall levels decreased across different road types from freeway>major arterial roads>minor arterial roads>collector roads>background sites.The ΔDF showed declines of 65%，55%，65% and 84% respectively for freeways，major arterial，minor arterial and collector roads from August 2007 to August 2008，and declines of 77%，76%，82% and 82% between August 2006 and August 2008.The ΔDF declined by 80%，79%，82% and 69% for freeways，major arterial，minor arterial and collector roads respectively between September 2007 and September 2008，and declined by 84%，88%，80% and 81% between September 2006 and September 2008.

Eighty samples were collected in August 2007 and August 2008 and analyzed for silt loading.PM10 emission factors and emission strengths were calculated using the AP-42 model.The silt loading reduced by 77%，35%，61%，59% and 75% for freeways，major arterial，minor arterial，collector and local roads respectively.The PM10 emission factors were reduced by 57%，15%，36%，51% and 61% and the PM10 emission strength declined by 70%，40%，55%，65% and 72% for freeways，major arterial，minor arterial，collector and local roads respectively between August 2007 and August 2008.The decline is consistent with the reduction in road dust-fall.

Key words：Paved road；Fugitive dust；Silt loading；Dust-fall；Olympic Games 2008；Beijing

1. Introduction

Long-term measurements have shown high levels of fine particulates（PM10）in the city of Beijing（Bergin et al.，2001；Chan et al.，2005）.Several recent studies have investigated the sources of PM10 in Beijing，which suggest fugitive road dust is one of the main sources（Song et al.，2006a，b）.Song et al.（2006a）used the CALPUFF modeling system to simulate PM10 dispersion from 1 January 2000 to 29 February 2000 in Beijing.There results suggest motor vehicles and road dust contributed 5% and 13%，respectively.Positive matrix factorization（PMF）to apportion sources of PM2.5 in Beijing，found the road dust contribution was 9%（Song et al.，2006b）.Dust emission rates from paved roads are a function of the silt loadings and particle size distribution of the road dust（Kuhn et al.，2001），vehicle speed（Etyemezian et al.，2003）and the size of the vehicle （Gillies et al.，2005）.Dust particles emitted by vehicular traffic consist of aggregates of fine clay to quartz particles and the size distribution can vary by several orders of magnitude（Pinnick et al.，1985）.

Several studies have investigated the efficiency of road fugitive dust control measures.Chang et al.（2005）studied the effectiveness of street sweeping and washing for controlling ambient TSP，and concluded that street sweeping followed by washing offered a measurable reduction in TSP emission potential.Typically，the reduction efficiency for ambient TSP was up to 30%.Gertler et al.（2006）performed roadside measurements of PM flux and conducted PM measurements from instrumented vehicles to evaluate the effectiveness of street sweeping in reducing dust re-entrainment and assess the impact of abrasives and de-icers on ambient PM near an alpine lake.They reported street sweeping to remove deposited material from the road surface resulted in an increase in the observed emission rate in short-term.

Normana and Johansson（2006）considered the effects of studded tires，sweeping and washing of paved surfaces and use of de-icing solutions on PM10 concentrations.They found that a 10% decrease in the fraction of studded tires used on the roads reduced the weekly average street canyon PM10 levels（due to local road abrasion） by about 10mg·m-3 ，although only daytime and dry street conditions were considered.Normana and Johansson（2006）also report that intense sweeping or washing of the pavement resulted in marginal reductions of PM10 levels（<10%）.We suggest that this control method would not have a significant influence on the PM10 levels in Beijing.

The Beijing government adopted strict control measures to reduce pollutant emission during the Olympic Games period，and the air quality showed a definite improvement during August 2008.The number of days the air quality fell within defined categories during the month of August in 2006，2007 and 2008 is shown in Fig.1.

A number of measures were taken to reduce fugitive road dust during the Olympic Games period（from 20 July to 20 September 2008）including：1）all construction sites were shut down and heavy trucks，used in construction，were taken off the road；2）road sweeping and washing were increased；3）motor vehicle use was restricted to alternate days based on odd and even license plate numbers.With these measures，the traffic volume of typical roads was reduced by 30% during the Olympic Games period compared with the pre-Games period.

2　Experimental methods

2.1　Road dust-fall monitors

This research used dust-fall gauges to monitor road dust-fall in Beijing.The size of the dust-fall gauges and the dust-fall analysis procedures were carried out in accordance with “Ambient air -Determination of dust-fall，Gravimetric method”（Ministry of Environmental Protection of the People’s Republic of China，1994）.The dust-fall gauges were constructed from 95# glass.

Monitoring sites were selected in accordance with the following principles：

（1）The gauges were mainly installed in green belts between main roads and side roads，although some were located on pavement.

（2）There were no construction sites within 100m of the monitoring site，and no shops，market places，parking lots or bus station within a 30m radius.

（3）The height of all the gauges was 3 m，and the sampling period was one month.

（4）The influences from traffic were minimized when sampling which ensures the safety of sample collectors.

Forty roads were selected to monitor dust-fall，and these were classified as freeways，major arterial，minor arterial or collector roads.The 40roads were distributed through eight areas of the city of Beijing，which were considered to be representative of the different areas that comprise the city（see Fig.2）.Two monitoring sites were established on each road，giving a total of 80road dustfall （DFr）gauges.Background monitoring sites were set up in 14 parks or greenbelt areas，giving 14 background dust-fall（DFb）gauges.In total 94 dust-fall measurements were made each month.

The difference between the road dust-fall（DFr）and background dust-fall（DFb）is called ΔDF in this paper，defined as the dust-fall attributable to that generated by the nearby road.

In this paper，data from April 2005 to December 2008 were used to characterize the DFr emissions during the 2008 Olympic Games.

2.2　Silt loading sampling

Silt loading is a key parameter for road fugitive dust emissions （EPA，2006a，b）.For this paper，80 sample sites were selected which included 16 freeways，20 major arterial roads，20minor arterial roads，16 collector roads and 8 local roads.These were sampled for dust loading in August 2007 and August 2008，and a total of 160 samples were obtained.The sampling and laboratory analysis procedures were carried out in accordance with Appendices C.1 and C.2 of the AP-42 document（EPA，2006a）.The sampling locations are shown in Fig.2.

2.3　PM10 emission calculations

Dust emissions from paved roads have been found to vary depending on the ‘silt loading’（sL）-the mass of the silt-sized material（≤75μm in diameter）per unit area of the travel surface-on the surface and the average weight of vehicles traveling on the road（EPA，2006b）.The quantity of dust emissions from movement of vehicles on a paved road can be estimated from：

　　（1）

where E is the emission rate of size specific PM（same units as “k”），sL is the silt load（g·m-2），W is the mean weight of the vehicle fleet （tons），and k is a constant（function of particle size）in g VKT-1 （vehicle kilometer traveled）（e.g.，k=4.6 for PM10）.Although there is some uncertainty in estimating fugitive road dust（FRD）emissions using this equation，it has been widely adopted（Limpaseni，2001）and is used in this paper.

FRD emission strength for a typical road varies with the emission factor and traffic volume，and can be calculated as：

ES=E×TV/1000　　（2）

where ES is the FRD emission strength（kg·km-1 day），E is the emission rate of size specific PM（g VKT-1），TV is the traffic volume for a typical road（day-1）.

The FRD emission quantity for a typical area（grid）varied with the length of different types of roads and emission strength of the roads，and can be calculated as：

　　（3）

where，EQ is the FRD emission quantity（kg·day-1），ESi is the FRD emission strength for road type i（kg·km-1 day），and Li is the length of road type i（km）.

3　Results and discussion

3.1　Road dust-fall

The average ΔDF values for each month from January 2006 to December 2008，are shown in Fig.3.We can see that the highest values occurred during summer and autumn 2006，while the values in spring and winter in 2006 were very similar to those in 2007.The values from January to June in 2008 were consistent with the same months in 2006 and 2007，but between June and September 2008 there was a sharp decline in the ΔDF values，with the lowest recorded value for the three year period occurring in September 2008.After September，the values increased.

The decline in road dust-fall was due to the control measures implemented during the 2008 Olympic Games in Beijing.

The road dust-fall（DFr）and background dust-fall（DFb）from January to December 2008，are shown in Fig.4.This figure shows that DFr is clearly larger than DFb，and DFr and DFb both have distinct monthly variations.The peak dust-fall values occurred in June，and based on the 2006 data，the values typically increase during summer.The results for 2008 indicated that the dust-fall level declined beginning in July because of the strict control measures implemented in Beijing.

The road dust-fall level in July declined by 37% compared with June，and declined by a further 38% in August when compared with July.The minimum road dust-fall value occurred in September.The ΔDF in September 2008 was 84% lower than in June 2008，and 76% and 82% lower than September 2007 and 2006，respectively.The background dust-fall level was 5.9t·km-2/30 day in August 2008，a decrease of 35% compared with August 2007.

The DFr for different types of roads and DFb for the background sites in the months of August and September from 2006 to 2008，are shown in Fig.5.Fromthese results we can see that DF decreases from freeways>major arterial roads>minor arterial roads>collector roads>background sites.The dust-fall levels during August and September 2008 clearly declined when compared with the same months in 2006 and 2007.

The ΔDF declined by 65%，55%，65% and 84% for freeways，major arterial，minor arterial and collector roads，respectively from August 2007 to August 2008，and declined by 77%，76%，82% and 82%，respectively from August 2006 to August 2008.The ΔDF declined by 80%，79%，82% and 69% for freeways，major arterial，minor arterial and collector roads，respectively from September 2007 to September 2008，and declined 84%，88%，80% and 81%，respectively from September 2006 to September 2008.

The frequency distribution of dust-fall levels at the 80 monitoring sites during August and September from 2006 to 2008，is shown in Fig.6.From this figure we can see that in August and September 2006 and 2007，the most frequently occurring values were in the range of 10～50t·km-2 /30 day while in August and September 2008 the levels were and mainly distributed in the 10～ 20t·km-2 /30 day range.

3.2　PM10 emission

During the Olympic Games period（from 20 July to 20 September 2008）all the construction sites were shut down and the heavy trucks used in construction were taken off the roads.Road sweeping and washing were increased.These measures act to reduce the silt loading.The silt loading on different types roads in the Beijing metropolitan area during August 2007 and 2008，is shown in Fig.7.From this we can see that silt loading in August 2008 was clearly lower than in August 2007.The silt loading was reduced by 77%，35%，61%，59% and 75% for freeways，major arterial，minor arterial，collector and local roads，respectively.

The results of a statistical analysis of traffic volumes at the peak hour on the sampled roads in the Beijing metropolitan area，is shown in Fig.8.The PM10 emission strength was calculated using the data in Fig.8，and the daily traffic volumes were calculated using the peak hour value and a peak factor for different types of roads.

Based on the silt loading results and traffic information，the PM10 emission factors for different types of roads were calculated using the AP-42 model，and the results are shown in Fig.8.The PM10 emission factors were reduced by 57%，15%，36%，51% and 61% for freeways，major arterial，minor arterial，collector and local roads，respectively in August 2008 compared with August 2007.

During the Olympic Games period motor vehicle use was restricted using odd and even license plate numbers.With only half the number of vehicles able to travel during on a given day during that period，the traffic volumes of typical roads were reduced by 70% compared with normal.The PM10 emission strengths in August 2007 and August 2008 were calculated and the results are shown in Fig.8.From this we can see that the PM10 emission strength decreased by 70%，40%，55%，65% and 72% for freeway，major arterial，minor arterial，collector and local roads respectively in August 2008 compared with August 2007.The level of decrease is consistent with the decrease in dust-fall levels.

The road fugitive dust PM10 emission distribution is shown in Fig.9.The spatial resolution of the grid 4km×4km，and from this we can determine the spatial distribution of road dust emissions.Road length，emission factor and PM10 emission strength for the different road types in each grid were calculated and GIS-based emission databases were established.From Fig.9 we can see that the PM10 emissions in August 2008 were smaller than in August 2007，emission strength was higher in urban area than in the suburbs，and emissions in the southern and eastern areas were higher than in the north and west.

4　Conclusions

Strict measures were adopted to reduce pollutant emissions during the Olympic Games period（from 20 July to 20 September 2008），and the air quality showed a definite improvement during that time.All the construction sites were shut down and heavy trucks used in construction were taken off the road while road sweeping and washing was increased.Motor vehicle use was restricted by using odd and even license plate numbers.

The ΔDF showed a significant decrease between 2006 and 2008.Dust-fall occurred with the following pattern：freeways>major arterial roads>minor arterial roads>collector roads>background sites.The ΔDF declined by 65%，55%，65% and 84% for freeways，major arterial，minor arterial and collector roads，respectively in August 2008 compared with August 2007，and declined by 77%，76%，82% and 82%，respectively between August 2006 and August 2008.The ΔDF declined by 80%，79%，82% and 69% for freeways，major arterial，minor arterial and collector roads，respectively in September 2008 compared with September 2007，and declined by 84%，88%，80% and 81% between September 2006 and September 2008.

The silt loading on the roads was reduced by 77%，35%，61%，59% and 75% for freeways，major arterial，minor arterial，collector and local roads，respectively.Based on application of the US EPA AP-42 method，PM10 emission factors reduced by 57%，15%，36%，51% and 61% and the PM10 emission strength declined by 70%，40%，55%，65% and 72% for freeways，major arterial，minor arterial，collector and local roads，respectively in August 2008 compared with August 2007.The decline in these parameters is consistent with the measured reduction in road dust-fall.

References

[1]　Bergin，M.H.，Cass，G.R.，Xu，J.，Fang，C.，Zeng，L.M.，Yu，T.，Salmon，L.G.，Kiang，C.S.，Tang，X.Y.，Zhang，Y.H.，Chameides，W.L.，2001.Aerosol radiative，physical，and chemical properties in Beijing during June 1999.J.Geophys.Res.106，17969-17980.

[2]　Chan，C.Y.，Xu，X.D.，Li，Y.S.，Wong，K.H.，Ding，G.A.，Chan，L.Y.，Cheng，X.H.，2005.Characteristics of vertical profiles and sources of PM2.5，PM10 and carbonaceous aerosols in Beijing.Atmos.Environ.39，5113-5124.

[3]　Chang，Y.M.，Chou，C.M.，Su，K.T.，Tseng，C.H.，2005.Effectiveness of street sweeping and washing for controlling ambient TSP.Atmos.Environ.39，891-1902.

[4]　EPA，2006a.Compilation of Air Pollutant Emission Factors.APPENDIX C.1 PROCEDURES FOR SAMPLING SURFACE/BULK DUST LOADING[R] .In：Stationary Point and Area Sources，vol.1.United States Environmental Protection Agency，MO：Midwest Research Institute.

[5]　EPA，2006b.Paved Roads.Available on the internet.http：//www.epa.gov/ttn/chief/ap42/ch13/final/c13s0201.pdfS（accessed June 2007）.

[6]　Etyemezian，V.J.，Kuhn，H.，Gillies，M.，Green，M.，Pitchford，M.，Watson，J.，2003.Vehicle based road dust emissions measurements（I）：methods and calibration.Atmos.Environ.37，4559-4571.

[7]　Gertler，A.，Kuhns，H.，Abu-Allabanb，M.，Damm，C.，Gillies，J.，Etyemezian，V.，Clayton，R.，Proffitt，D.，2006.A case study of the impact of winter road sand/salt and street sweeping on road dust re-entrainment.Atmos.Environ.40，5976-5985.

[8]　Gillies，J.，Etyemezian，V.，Kuhn，H.，Nikolic，D.，Gillette，D.A.，2005.Effect of vehicle characteristics on unpaved road dust emissions.Atmos.Environ.39，2341-2347.

[9]　Kuhn，H.，Etyemezian，V.，Landwehr，D.，Mac-Dougall，C.，Pitchford，M.，Green，M.，2001.Testing re-entrained aerosol kinetic emissions from roads（TRAKER）：a new approach to infer silt loading on roadways.Atmos.Environ.35，2815-2825.

[10]　Limpaseni，W.，2001.Control of Road Dust.Regional Workshop on Fighting Urban Air Pollution：From Plan to Action.United Nations Conference Center，Bangkok，Thailand.

[11]　Ministry of Environmental Protection of the People’s Republic of China，1994.Ambient air - determination of dust-fall，Gravimetric method.http：//www.mep.gov.cn/image20010518/5325.pdf.

[12]　Normana，M.，Johansson，C.，2006.Studies of some measures to reduce road dust emissions from paved roads in Scandinavia.Atmos.Environ.40，6154-6164.

[13]　Pinnick，R.G.，Fernandez，G.，Hinds，B.D.，Bruce，C.W.，Schaefer，R.W.，1985.Dust generated by vehicular traffic on unpaved roadways：size and infrared extinction.Aerosol Sci.Technol.4，99-121.

[14]　Song，Y.，Zhang，M.，Cai，X.，2006a.PM10 modeling of Beijing in the winter.Atmos.Environ.40，4126-4136.

[15]　Song，Y.，Zhang，Y.，Xie，S.D.，Zhang，Y.H.，Wei，Y.J.，Zhang，M.S.，Zeng，L.M.，Lu，S.H.，2006b.Source apportionment of PM2.5 in Beijing by positive matrix factorization.Atmos.Environ.40，1526-1537.