Differences Between Weekday and Weekend Levels of Ozone,NO 2,NO x,and Respirable Suspended Particulates in Hong Kong

Abstract

Differences in ozone (O₃), nitrogen oxides (NO_x), and respirable suspended particulate (RSP) weekday and weekend concentrations in Hong Kong were investigated. Data were collected hourly by the Hong Kong Environmental Protection Department from January 1999 to October 2007 at 14 sites: 9 urban, 4 newly developed, and 1 rural station. Weekend O₃ concentrations were 14% and 22% higher for the afternoon peak O₃ time and 8-h O₃ average concentrations, respectively, at six urban stations, whereas NO₂ and NO_x concentrations were 19% and 25% lower, respectively, during the afternoon O₃ peak time on weekends. No significant differences in peak afternoon O₃ concentrations were found at the newly developed or rural stations. Weekend RSP concentrations were lower at most urban stations. Urban RSP concentrations were lower on weekends than weekdays, by 14% for the morning rush hour and 8% for the daily average RSP. O₃ formation sensitivity to volatile organic compounds and lower weekend NO_x emissions could explain the O₃ weekend effect in urban Hong Kong. Lower weekend RSP concentrations are predominantly the result of fewer anthropogenic emissions, primarily due to decreased traffic flow.

Introduction

Human activities have affected air quality over the past two decades. Daily variation in air pollution levels has been studied, because human activities vary between weekdays and weekends (Madhavi Latha and Badarinath, 2003; Riga-Karandinos et al., 2006). The weekend ozone (O₃) effect is generally characterized by higher O₃ concentrations on weekends than weekdays, despite relatively low O₃ precursor concentrations, such as nitrogen oxides (NO_x), including NO and NO₂, and volatile organic compounds (VOCs). This phenomenon was first reported in the United States in the 1970s (Cleveland et al., 1974; Lebron, 1975) and has been recognized in urban centers worldwide (Pont and Fontan, 2001; Jenkin et al., 2002; Qin et al., 2004; Blanchard and Tanenbaum, 2006; Sadanaga et al., 2008). Some studies have found lower respirable suspended particulate (RSP; i.e., particulate matter with an aerodynamic diameter of <10 μm) concentrations in urban centers on weekends (Qin et al., 2004; Jones et al., 2008), which could be caused by differences in anthropogenic emissions between weekdays and weekends.

Numerous large Chinese cities have experienced severe air pollution with the recent surge in economic growth. However, few studies have examined differences in weekday and weekend air pollution levels. This study investigated weekday and weekend O₃, NO₂, NO_x, and RSP pollution levels in Hong Kong. O₃ and RSP weekend effects were observed in urban centers, especially in downtown areas, where traffic contributes greatly to NO_x and RSP emissions and varies more between weekdays and weekends. The basic mechanisms of O₃ formation are as follows: \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland,xspace}\usepackage{amsmath,amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*} { \rm NO}_2 + hv \rightarrow { \rm NO} + { \rm O} ( ^3{ \rm P} ) \tag{1} \end{align*} \end{document} \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland,xspace}\usepackage{amsmath,amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*} { \rm O} ( ^3{ \rm P} ) + { \rm O}_2 + { \rm M} \rightarrow { \rm O}_3 + { \rm M} \tag{2} \end{align*} \end{document} \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland,xspace}\usepackage{amsmath,amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*} { \rm NO} + { \rm O}_3 \rightarrow { \rm NO}_2 + { \rm O}_2 \tag{3} \end{align*} \end{document}

Patterns in O₃ formation, destruction, and accumulation become more complex with other interactions. Chan et al. (1998) suggested that the prevailing air mass from the north industrial Pearl River Delta region is rich in O₃ precursors and could contribute greatly to the elevated O₃ levels in Hong Kong. Lau et al. (2008) observed lower NO₂, NO_x, and RSP emissions on Sundays and a positive correlation between traffic levels and NO₂, NO_x, and RSP concentrations in urban areas. Further, industrial activities, such as construction, may decrease on weekends, further reducing weekend RSP concentrations.

Data and Methods

Hong Kong is an international metropolis on the south coast of China with a subtropical monsoon climate. It has a population of about 7,026,400 in an area of about 1,104 km². Hourly data, collected from January 1990 to October 2007 downloaded from the Hong Kong Environment Protection Department website, were used for this study (http://epic.epd.gov.hk/ca/uid/cairdata/p/1). The Hong Kong Air Quality Monitoring Network comprised 14 fixed monitoring stations (Fig. 1): Causeway Bay (CB), Central (C), Central Western (CW), Eastern (E), Kwai Chung (KC), Kwon Tong (KT), Mong Kok (MK), Sham Shui Po (SSP), Tsuen Wan (TW), Sha Tin (ST), Tai Po (TP), Tung Chung (TC), Yuen Long (YL), and Tap Mun (TM).

FIG. 1.

Distribution of the fourteen monitoring stations in Hong Kong, where (Ro) stands for Roadside Station; (U) stands for Urban Station; (N) stands for New Town Station; (Ru) stands for Rural Station.

Characteristics of the pollutants monitored at each station are listed in Table 1. The stations are classified into three groups, based on location: nine urban, including three roadside stations, four newly developed stations, and one rural station. The urban stations reflect urban characteristics, including densely populated residential areas with commercial and industrial development. The newly developed stations reflected air pollution in new residential areas with rapid development close to urban centers. The rural station was on an outlying island 15 km from the nearest developed area and represents background pollution levels in Hong Kong. Sulfur dioxide was measured using UV fluorescence (TECO Model 43A; Monitor Laboratories 8850), NO₂ with a chemiluminescence analyzer (API 200A; Monitor Laboratories 8840), CO using nondispersive infrared absorption with gas filter correlation (TECO Model 48, 48C), and RSP with a gravimetric or oscillating microbalance (Graseby Andersen PM10 R&P TEOM Series 1400a-PM10).

Table 1.

Information of Monitoring Stations in Hong Kong Environment Protection Department

Site ^a	Type (height above ground)	Characteristics of station location areas	Pollutant measured
CB	Roadside (3 m)		NO₂, NO_x, O₃, RSP
C	Roadside (4.5 m)	Urban roadside with heavy traffic	NO₂, NO_x, O₃, RSP
MK	Roadside (3 m)		NO₂, NO_x, O₃, RSP
CW	Urban (18 m)	Residential area	NO₂, NO_x, O₃, RSP
E	Urban (15 m)	Densely populated residential area	NO₂, O₃, RSP
KC	Urban (13 m)	Densely populated residential area with mixed commercial/industrial developments	NO₂, NO_x, O₃, RSP
KT	Urban (25 m)		NO₂, NO_x, O₃, RSP
TW	Urban (17 m)		NO₂, NO_x, O₃, RSP
SSP	Urban (17 m)	Densely populated residential area with commercial developments	NO₂, NO_x, O₃, RSP
ST	New town (21 m)	Residential area	NO₂, NO_x, O₃, RSP
TP	New town (25 m)		NO₂, O₃, RSP
TC	New town (21 m)		NO₂, NO_x, O₃, RSP
YL	New town (25 m)	Residential areas with fairly rapid development	NO₂, NO_x, O₃, RSP
TM	Rural (11 m)	Outlying island as background station of Hong Kong	NO₂, NO_x, O₃, RSP

Monitoring station names are abbreviated: Causeway Bay (CB), Central (C), Central Western (CW), Eastern (E), Kwai Chung (KC), Kwon Tong (KT), Mong Kok (MK), Sham Shui Po (SSP), Tsuen Wan (TW), Sha Tin (ST), Tai Po (TP), Tung Chung (TC), Yuen Long (YL), and Tap Mun (TM).

RSP, respirable suspended particulate.

O₃, its NO_x precursors, and RSP were selected for evaluation in this study based on previous studies (Qin et al., 2004; Blanchard and Tanenbaum, 2006; Murphy et al., 2007). The original average hourly data were separated into weekdays, Monday—Saturday, and weekends, Sunday. Sundays alone were treated as weekends, because before July 1, 2007, Saturday mornings were regular workdays. The means and standard deviations for the 24-h periods were calculated. The mean pollution levels between weekdays and weekends were analyzed with t-tests using a 5% significance level.

Three statistical comparisons were used to examine O₃ and RSP concentrations between weekdays and weekends. First, the concentrations of the four pollutants at 09:00 local time were compared. Since this time was in the morning rush hour, pollutant concentrations reflected emissions primarily from vehicle exhausts (Qin et al., 2004; Blanchard and Tanenbaum, 2006). Second, O₃, NO₂, and NO_x concentrations at the peak afternoon O₃ time and maximum 8-h average O₃ concentrations were compared to examine weekend O₃ effects (Qin et al., 2004; Blanchard and Tanenbaum, 2006). Finally, weekday and weekend daily average RSP concentrations were compared, because the levels of some primary RSPs, such as road and construction dust, may decrease on weekends (Lonati et al., 2006).

For each monitoring station, the specified air pollutant, and time of day, quantitative differences between the mean weekday and weekend concentrations were expressed as the percentage decrease from the weekend to weekday mean (Murphy et al., 2007): \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland,xspace}\usepackage{amsmath,amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*} { \rm Difference} = & 100 \times ( \hbox{Weekend mean} - \hbox{Weekday mean} ) \\ & / ( \rm \hbox{Weekday mean} ) \end{align*} \end{document}

Results and Discussion

Statistical comparisons

Statistical comparisons for the four pollutants at each monitoring station are shown in Table 2. O₃ concentrations were not monitored at the roadside stations (CB, C, and MK). Air pollutant concentrations monitored at urban stations differed significantly between weekdays and weekends, except RSP during the morning rush hour and the daily average concentration at station E.

Table 2.

Statistical Results of the Differences for O₃, NO₂, NO_x, and Respirable Suspended Particulate Concentrations Between Weekday and Weekend at the Fourteen Monitoring Stations

		Morning ^b				Afternoon ^c
Site ^a	Pollutant (μg/m ³ )	Weekday ^d	Sunday ^d	Difference ^e	Scale	Weekday ^d	Sunday ^d	Difference ^e	Scale
CB	NO₂	100±31 (2,617)^d	88±28 (442)	Y^e	−13%	112±37 (2,546)	104±35 (442)	Y	−7%
	NO_x	472±212 (2,617)	432±205 (442)	Y	−9%	433±175 (2,546)	414±184 (442)	Y	−4%
	RSP	95±36 (2,647)	76±36 (447)	Y	−20%	105±38 (2,519)	96±36 (445)	Y	−9%
	RSP (daily)	90±28 (2,714)	85±28 (451)	Y	−5%
C	NO₂	108±33 (2,734)	83±28 (451)	Y	−24%	117±41 (2,594)	95±37 (455)	Y	−19%
	NO_x	521±196 (2,734)	290±129 (451)	Y	−44%	415±146 (2,594)	272±110 (455)	Y	−34%
	RSP	86±35 (2,696)	66±32 (450)	Y	−23%	79±36 (2,496)	65±32 (454)	Y	−18%
	RSP (daily)	73±29 (2,698)	64±28 (454)	Y	−13%
MK	NO₂	97±28 (2,105)	88±27 (350)	Y	−9%	115±37 (2,069)	108±35 (348)	Y	−6%
	NO_x	434±106 (2,105)	347±99 (350)	Y	−20%	359±83 (2,069)	346±73 (348)	Y	−4%
	RSP	74±31 (2,104)	64±30 (353)	Y	−14%	78±35 (2,043)	67±32 (352)	Y	−15%
	RSP (daily)	69±27 (2,118)	63±28 (353)	Y	−9%
CW	O₃	22±17 (2,625)	30±23 (440)	Y	34%	60±41 (2,633)	66±39 (436)	Y	9%
	O₃ (max 8-h)	38±25 (2,573)	46±28 (437)	Y	19%
	NO₂	59±26 (2,605)	48±23 (440)	Y	−20%	54±29 (2,621)	43±26 (432)	Y	−21%
	NO_x	126±89 (2,626)	89±77 (440)	Y	−29%	76±42 (2,621)	57±33 (432)	Y	−26%
	RSP	55±34 (2,623)	50±32 (441)	Y	−9%	57±35 (2,546)	51±33 (440)	Y	−12%
	RSP (daily)	53±29 (2,640)	49±30 (439)	Y	−6%
E	O₃	25±14 (2,455)	31±17 (414)	Y	22%	51±27 (2,538)	57±28 (423)	Y	11%
	O₃ (max 8-h)	36±19 (2,451)	42±21 (422)	Y	17%
	NO₂	64±23 (2,457)	53±23 (414)	Y	−17%	59±23 (2,521)	48±20 (421)	Y	−19%
	RSP	49±29 (2,650)	46±29 (444)	N	−6%	51±30 (2,577)	46±29 (440)		−10%
	RSP (daily)	47±26 (2,691)	45±26 (445)	N	−5%
KC	O₃	13±12 (2,728)	20±17 (457)	Y	52%	38±30 (2,744)	47±33 (455)	Y	23%
	O₃ (max 8-h)	24±19 (2,715)	32±23 (455)	Y	34%
	NO₂	72±25 (2,729)	55±20 (460)	Y	−25%	74±33 (2,734)	56±30 (455)	Y	−24%
	NO_x	252±132 (2,729)	146±81 (460)	Y	−42%	160±89 (2,734)	104±57 (455)	Y	−35%
	RSP	59±30 (2,714)	50±29 (459)	Y	−16%	60±33 (2,670)	52±32 (454)	Y	−13%
	RSP (daily)	56±26 (2,746)	51±26 (456)	Y	−10%
KT	O₃	16±13 (2,471)	23±17 (415)	Y	41%	45±29 (2,488)	51±33 (413)	Y	14%
	O₃ (max 8-h)	30±20 (2,446)	36±23 (412)	Y	21%
	NO₂	75±27 (2,477)	61±24 (414)	Y	−19%	72±30 (2,479)	61±27 (408)	Y	−16%
	NO_x	234±105 (2,477)	155±94 (414)	Y	−34%	143±62 (2,479)	114±53 (408)	Y	−21%
	RSP	61±31 (2,500)	53±31 (418)	Y	−13%	60±31 (2,467)	52±30 (417)	Y	−14%
	RSP (daily)	55±27 (2,538)	51±27 (417)	Y	−9%
TW	O₃	15±12 (2,575)	20±16 (432)	Y	37%	46±37 (2,579)	50±35 (429)	Y	9%
	O₃ (max 8-h)	29±22 (2,537)	34±24 (429)	Y	17%
	NO₂	68±24 (2,569)	56±22 (429)	Y	−17%	70±32 (2,565)	61±35 (426)	Y	−13%
	NO_x	187±95 (2,571)	132±81 (429)	Y	−29%	120±56 (2,566)	98±48 (426)	Y	−18%
	RSP	58±31 (2,569)	52±30 (431)	Y	−10%	63±37 (2,509)	57±36 (432)	Y	−10%
	RSP (daily)	56±28 (2,589)	53±28 (430)	Y	−6%
SSP	O₃	13±11 (2,639)	18±15 (439)	Y	42%	41±34 (2,658)	47±34 (438)	Y	15%
	O₃ (max 8-h)	26±20 (2,631)	32±23 (438)	Y	26%
	NO₂	70±25 (2,635)	59±24 (439)	Y	−16%	71±28 (2,645)	57±25 (436)	Y	−20%
	NO_x	188±92 (2,634)	136±83 (439)	Y	−27%	125±47 (2,645)	96±45 (436)	Y	−23%
	RSP	56±31 (2,650)	50±30 (441)	Y	−11%	59±32 (2,621)	52±31 (441)	Y	−12%
	RSP (daily)	55±27 (2,667)	51±27 (439)	Y	−8%
ST	O₃	25±23 (2,664)	29±23 (444)	Y	15%	74±46 (2,670)	78±46 (442)	N	5%
	O₃ (max 8-h)	49±32 (2,651)	55±32 (442)	Y	11%
	NO₂	52±24 (2,676)	43±23 (445)	Y	−18%	36±21 (2,662)	30±18 (440)	Y	−17%
	NO_x	125±113 (2,676)	92±90 (445)	Y	−27%	50±33 (2,662)	39±26 (440)	Y	−21%
	RSP	53±32 (2,648)	47±30 (442)	Y	−11%	52±32 (2,589)	47±31 (439)	Y	−10%
	RSP (daily)	51±27 (2,685)	47±27 (445)	Y	−7%
TP	O₃	30±19 (2,380)	33±20 (393)	Y	12%	73±39 (2,435)	74±38 (395)	N	1%
	O₃ (max 8-h)	52±28 (2,366)	54±27 (396)	Y	6%
	NO₂	57±25 (2,412)	46±22 (399)	Y	−19%	41±21 (2,439)	36±18 (395)	Y	−13%
	RSP	55±34 (2,465)	51±31 (409)	Y	−8%	52±32 (2,406)	50±33 (404)	N	−3%
	RSP (daily)	51±27 (2,498)	49±28 (408)	N	−4%
TC	O₃	28±21 (2,659)	31±22 (443)	Y	11%	81±60 (2,659)	85±57 (440)	N	4%
	O₃ (max 8-h)	49±33 (2,593)	55±33 (440)	Y	12%
	NO₂	45±25 (2,646)	40±24 (438)	Y	−11%	45±36 (2,637)	38±32 (433)	Y	−14%
	NO_x	86±64 (2,646)	70±56 (438)	Y	−18%	58±47 (2,637)	48±40 (433)	Y	−17%
	RSP	51±35 (2,627)	49±34 (435)	N	−4%	63±47 (2,495)	58±47 (433)	Y	−8%
	RSP (daily)	52±33 (2,645)	50±33 (435)	N	−4%
YL	O₃	19±15 (2,612)	25±17 (442)	Y	27%	65±48 (2,672)	69±45 (443)	N	7%
	O₃ (max 8-h)	41±29 (2,585)	47±30 (442)	Y	14%
	NO₂	61±24 (2,615)	48±23 (444)	Y	−21%	57±30 (2,671)	46±27 (443)	Y	−19%
	NO_x	146±97 (1,214)	101±71 (205)	Y	−30%	86±43 (1,216)	71±40 (201)	Y	−17%
	RSP	66±39 (2,672)	61±39 (456)	Y	−8%	66±42 (2,624)	62±44 (456)	N	−6%
	RSP (daily)	60±33 (2,726)	57±33 (457)	Y	−6%
TM	O₃	54±29 (2,637)	52±27 (416)	N	−5%	104±46 (2,547)	102±47 (416)	N	−1%
	O₃ (max 8-h)	75±35 (2,494)	74±33 (416)	N	−2%
	NO₂	15±13 (2,605)	16±14 (426)	N	5%	9.2±8.2 (2,521)	9.4±8.2 (410)	N	2%
	NO_x	21±22 (2,605)	22±21 (426)	N	4%	11.6±9.4 (2,521)	11.9±9.3 (410)	N	2%
	RSP	48±30 (2,653)	46±29 (441)	N	−4%	49±31 (2,485)	47±31 (447)	N	−4%
	RSP (daily)	46±26 (2,686)	44±26 (448)	N	−3%

Monitoring station names are abbreviated as in Table 1.

09:00 (Local Time) for NO₂, NO_x, O₃, and RSP.

Peak ozone time for O_3, NO₂, and NO_x; 14:00 (Local Time) for RSP.

Mean±standard deviation (data number).

Y stands for significant difference between weekday and weekend, and N stands for no significant difference between weekday and weekend, 5% level of significance is selected in this study.

The four newly developed stations exhibited no significant difference in afternoon peak O₃ concentrations, and there was no significant difference between weekday and weekend daily average RSP concentrations at TP and TC and morning RSP concentrations at TC.

No significant difference was found at the rural station, TM, for the four pollutants monitored.

Weekday/weekend differences

Table 2 summarizes the differences between the weekday and weekend peak O₃ and maximum 8-h average O₃ concentrations, NO₂, NO_x, and RSP concentrations during the morning rush hour, NO₂ and NO_x concentrations during the afternoon peak O₃ time, RSP concentrations at 14:00, and daily average RSP concentrations at the 14 stations.

The peak O₃ and maximum 8-h O₃ average concentrations were higher on weekends than weekdays at all stations except TM, whereas the six urban stations, apart from the three roadside stations, had significantly different O₃ concentrations, at 14% and 22% higher for the afternoon peak O₃ time and maximum 8-h O₃ average. The difference was greater for the maximum 8-h O₃ average than the afternoon peak O₃ concentration, consistent with previous studies (Marr and Harley, 2002; Qin et al., 2004); the 8-h metric appears to be more sensitive to daily O₃ concentration differences.

The concentrations of NO₂ and NO_x in the morning rush hour and afternoon peak O₃ time were significantly lower on weekends than weekdays at all stations except TM. The concentrations at the six urban stations, which showed significant O₃ concentration increases, were 19% and 25% lower on weekends than weekdays during the afternoon peak O₃ time.

The morning rush hour, afternoon peak O₃ time, and daily average RSP concentrations were lower on weekends than weekdays at all 14 stations. The nine urban stations revealed significant RSP decreases, of 14% and 8% for the morning rush hour and daily average, respectively, from weekdays to weekends.

The discussion of the O₃ and RSP weekend effects has been organized by the urban, newly developed, and rural locations, because spatial variation was observed. Previous studies (Lam et al., 2005) have shown that O₃ formation is limited by VOCs in urban Hong Kong and surrounding areas. However, VOC monitoring was not included in this study. We refer to previous studies of the relationship between VOCs and O₃ to discuss O₃ formation and weekend effects.

Differences among urban stations

A relatively high O₃weekend effect was observed in the urban stations CW, E, KC, KT, SSP, and TW.

At the nine urban stations, the NO_x concentrations at rush hour and afternoon peak O₃ time decreased significantly from weekdays to weekends. Figure 2a and b illustrate the daily average difference in weekdays and weekends in NO_x and O₃ at KC, respectively; the other urban stations followed similar trends. Significant NO_x decreases directly reduce NO_x inhibition of O₃ formation (So and Wang, 2003; Wang and Lu, 2006). O₃ accumulation begins at 08:00 on weekends and 09:00 on weekdays. Since less NO is emitted, there is less reaction of O₃ with NO (i.e., NO+O₃→NO₂+O₂). Further, fewer hydroxyl radicals are removed, because there is less NO₂ present (NO+OH→HNO₃; Lawson, 2003).

FIG. 2.

(a), (b), and (c) showed average daily NO_x, O₃ and RSP variations between weekday and Sunday at KC, respectively; (d), (e), and (f) showed average daily NO_x, O₃ and RSP variations between weekday and Sunday at ST, respectively. RSP, respirable suspended particulate.

The greatest contributors of NO_x emissions are public electricity generation and road transportation (www.epd.gov.hk/epd/english/environmentinhk/air/data/emission_inve06.html, Hong Kong Air Pollutant Emission Inventory). Since public electricity generation differs little between weekdays and weekends, decreased traffic flow is likely the predominant cause of the lower NO_x concentrations on weekends (Lau et al., 2008). Another hypothesis involves the decreased fine particles emissions. We observed significant RSP concentration decreases during the afternoon peak O₃ time at the urban stations on weekends, which would increase the radiation available for photolysis, increasing the O₃ formation rate (Lawson, 2003).

During the day, weekend RSP concentrations were lower at all urban stations. Figure 2c illustrates the daily RSP variation at KC. From 08:00 to 18:00, significant reductions in weekend RSP concentrations were observed. The decrease was particularly large during the morning rush hour and at roadside stations. Road transportation and public electricity generation are the top emission sources (HIAPEI); Further, Lau et al. (2008) found a positive correlation between traffic flow and RSP concentrations in urban areas. Thus, reduced vehicular particulate matter emissions and roadway transport dust could be the primary cause of the RSP weekend effect.

Differences in new town stations

There was no great difference between weekday and weekend O₃ levels at the four newly developed stations.

Figure 2d and e show the daily average NO_x and O₃ variation between weekdays and weekends at ST; similar trends were observed at the other newly developed stations. Although the four stations experienced large NO_x decreases on weekends, the afternoon peak O₃ concentration was not significantly different between weekdays and weekends. These stations were downwind of the Pearl River Delta and O₃ formation could have been influenced by regional air masses from Mainland China (Chan et al., 1998; Wang and Lu, 2006). However, the 8-h maximum average O₃ concentrations at these stations differed significantly between weekdays and weekends, suggesting that, with rapid new development, human activities and local influences could contribute increasingly to O₃ levels.

The newly developed stations generally exhibited smaller decreases in RSP concentrations between weekdays and weekends than the urban stations, because human activities were not so intensive. However, there were significant differences between the weekday and weekend RSP concentrations at ST and YL (e.g., Fig. 2f), because they are located in residential areas experiencing rapid development and there is more anthropogenic interference.

Differences in rural stations

There were no significant differences between the weekday and weekend air pollutant concentrations at TM, because there is little human activity there and it is downwind of Mainland China. It is influenced primarily by regional emissions and chemical and transport processes (So and Wang, 2003).

Moreover, the temporal patterns of the O₃, NO₂, and NO_x concentrations at TM were the opposite of the other stations. Lower weekend O₃ concentrations at the afternoon peak O₃ time and maximum 8-h O₃ average were observed at TM, consistent with previous studies (Cleveland and McRae, 1978; Altshuler et al., 1995; Heuss et al., 2003).

Conclusions

Urban Hong Kong monitoring stations showed significant differences in O₃, NO₂, NO_x, and RSP concentrations between weekdays and weekends from January 1999 to October 2007. The afternoon peak O₃ and maximum 8-h average O₃ concentrations at six urban stations with significant O₃ differences were 14% and 22% higher on weekends, respectively. In contrast, the NO₂ and NO_x concentrations during the afternoon peak O₃ time were 19% and 25%, respectively, lower on weekends. The nine urban stations revealed RSP decreases of 14% and 8% for the morning rush hour and the daily average, respectively, from weekdays to weekends.

Combined with a weekend NO_x emission decrease, VOC-sensitivity could explain the O₃ weekend effect in urban Hong Kong. Further, lower RSP concentrations could reduce sunlight scattering and enhance O₃ formation on weekends. Relatively lower morning O₃ precursor emissions on weekends did not produce significant O₃ weekend effects at the newly developed stations. This could be explained by influences from local and long-range transported emissions. The rural site exhibited no significant difference for the four pollutants, because it received no local input and was influenced only by regional emissions, chemicals, and transport processes from Mainland China.

Footnotes

Acknowledgment

The hourly air pollution data were downloaded from the internal website of the Hong Kong Environment Protection Department.

Author Disclosure Statement

No competing financial interests exist.

References

Altshuler

S.L.

et al. 1995. Weekday vs. weekend ambient ozone concentrations: discussion and hypotheses with focus on northern California. J. Air Waste Manage. Assoc., 45:967.

Blanchard

C.L.

, Tanenbaum

2006. Weekday/weekend differences in ambient air pollutant concentrations in Atlanta and the Southeastern United States. J. Air Waste Manage. Assoc., 56:271.

Chan

L.Y.

et al. 1998. Surface ozone pattern in Hong Kong. J. Appl. Meteorol., 37:1153.

Cleveland

W.S.

et al. 1974. Sunday and workday variations in photochemical air pollutants in New Jersey and New York. Science, 186:1037.

Cleveland

W.S.

, Mcrae

J.E.

1978. Weekday-weekend ozone concentrations in the Northeast United States. Environ. Sci. Technol., 12:558.

Heuss

J.M.

et al. 2003. Weekday/weekend ozone differences: what can we learn from them? J. Air Waste Manage. Assoc., 53:772.

Jenkin

M.E.

et al. 2002. The origin and day-of-week dependence of photochemical ozone episodes in the UK. Atmos. Environ., 36:999.

Jones

A.M.

et al. 2008. The weekday-weekend difference and the estimation of the non-vehicle contributions to the urban increment of airborne particulate matter. Atmos. Environ., 42:4467.

Lam

K.S.

et al. 2005. Study on an ozone episode in hot season in Hong Kong and transboundary air pollution over Pearl River Delta region of China. Atmos. Environ., 39:1967.

10.

Lau

et al. 2008. Contributions of roadside vehicle emissions to general air quality in Hong Kong. Transp. Res. Part D Transp. Environ., 13:19.

11.

Lawson

D.R.

2003. The weekend ozone effect-The weekly ambient emissions control experiment. Environ. Manage., 32:17.

12.

Lebron

1975. A comparison of weekend-weekday ozone and hydrocarbon concentrations in the Baltimore-Washington metropolitan area. Atmos. Environ., 9:861.

13.

Lonati

et al. 2006. The role of traffic emissions from weekends' and weekdays' fine PM data in Milan. Atmos. Environ., 40:5998.

14.

Madhavi Latha

K.M.

, Badarinath

K.V.S.

2003. Black carbon aerosols over tropical urban environment—a case study. Atmos. Res., 69:125.

15.

Marr

L.C.

, Harley

R.A.

2002. Spectral analysis of weekday-weekend differences in ambient ozone, nitrogen oxide, and non-methane hydrocarbon time series in California. Atmos. Environ., 36:2327.

16.

Murphy

J.G.

et al. 2007. The weekend effect within and downwind of Sacramento— Part 1: observations of ozone, nitrogen oxides, and VOC reactivity. Atmos. Chem. Phys., 7:5327.

17.

Pont

, Fontan

2001. Comparison between weekend and weekday ozone concentration in large cities in France. Atmos. Environ., 35:1527.

18.

Qin

et al. 2004. Weekend/weekday differences of ozone, NOx, CO, VOCs, PM10 and the light scatter during ozone season in southern California. Atmos. Environ., 38:3069.

19.

Riga-Karandinos

A.-N.

et al. 2006. Study of the weekday–weekend variation of air pollutants in a typical Mediterranean coastal town. Int. J. Environ. Pollut., 27:300.

20.

Sadanaga

et al. 2008. Weekday/weekend difference of ozone and its precursors in urban areas of Japan, focusing on nitrogen oxides and hydrocarbons. Atmos. Environ., 42:4708.

21.

K.L.

, Wang

2003. On the local and regional influence on ground-level ozone concentrations in Hong Kong. Environ. Pollut., 123:307.

22.

Wang

, Lu

W.Z.

2006. Forecasting of ozone level in time series using MLP model with a novel hybrid training algorithm. Atmos. Environ., 40:913.