Weekday/weekend difference of ozone and its precursors in urban areas of Japan, focusing on nitrogen oxides and hydrocarbons

Abstract

This paper analyzes the weekday–weekend differences of ozone, NO_x (NO and NO₂) and non-methane hydrocarbons (NMHCs) in Tokyo and Osaka, Japan, from 2001 to 2004, in order to investigate the ozone weekend effect (OWE). Data were measured at 24 and 23 general air monitoring stations in Tokyo and Osaka, respectively. We focused on NO_x and NMHCs to investigate the causes of OWE. Specifically, we examined weekday/weekend differences in the titration of O₃ by NO and the photochemical ozone production. Weekend ozone concentrations were confirmed to be greater than those of weekdays during most periods in both Tokyo and Osaka, although precursor concentrations were higher on weekdays. Weekday ozone concentrations in Tokyo (other than during 2003) and Osaka decreased due to the reaction of O₃ with NO, resulting in higher ozone but lower O_x (the sum of O₃ and NO₂) concentrations on weekends than on weekends. In 2003 in Tokyo, however, the effect of the photochemical ozone production on weekends was larger than on weekdays, as determined by the comparison of ozone concentration increase versus NMHCs/NO_x ratio and O_x between weekdays and the weekend. The weekend NMHCs/NO_x ratio would be in the boundary area between volatile organic compound (VOC)- and NO_x-limited regime, so it is considered that the difference between 2003 and the other periods in Tokyo might be due to a change in precursor limitation.

Introduction

Recently, ozone concentrations in the urban atmosphere have been rising despite a steady decrease in the concentrations of ozone precursors such as NO_x (NO+NO₂) and volatile organic compounds (VOCs). In Japan, for example, these trends were observed by the recent continuous measurements at the many air monitoring stations in the Tokyo Metropolitan area. The cause of the recent increase in the air pollution of ozone (e.g., Chou et al., 2006) remains unclear.

The ozone “weekend effect” (OWE) is a common phenomenon of ozone behavior in the urban atmosphere: ozone concentrations on weekends are higher than those on weekdays despite lower concentrations of ozone precursors. The OWE was discovered in the 1970s in the United States; more recently, the same phenomenon was recognized in several other countries as well (e.g., Diem, 2000; Jenkin et al., 2002; Marr and Harley, 2002; Qin et al., 2004a, Qin et al., 2004b; Paschalidou and Kassomenos, 2004; Jiménez et al., 2005; Gao et al., 2005; Riga-Karandinos and Saitanis, 2005). However, little research has been conducted in Japan and the CARB (California Air Resource Board, 2001) hypotheses will be examined. The OWE and the recent ozone increase in the urban atmosphere have both occurred despite lowered levels of ozone precursors, so the research on the OWE might relate to ozone control strategies in the urban air.

The behavior of ozone in the urban atmosphere is very complex. Ozone concentrations are influenced by various meteorological factors such as solar radiation, temperature, wind direction and velocity. Ozone concentrations are also intricately related to NO_x and VOCs concentrations, the chemical precursors of ozone. Fig. 1 shows the fundamental reaction scheme for photochemical ozone production in the urban atmosphere. Ozone in the troposphere is produced in the photolysis of NO₂ with solar UV (λ<420 nm):

NO₂+hν→NO+O(³P),

O(³P)+O₂+M→O₃+M,

where M represents a third-body molecule. These reactions have a following inverse reaction to regenerate NO₂ and O₂:

NO+O₃→NO₂+O₂.

Substantially, the photochemical ozone production is determined by the reaction of NO with peroxy radicals:

RO₂+NO→RO+NO₂,

RO+O₂→R′COR″+HO₂,

HO₂+NO→OH+NO₂,

where R, R′ and R″ indicate alkyl groups. The peroxy radicals are generated by the reaction of OH with VOCs such as non-methane hydrocarbons (NMHCs):

OH+NMHCs→R+H₂O,

R+O₂+M→RO₂+M.

As shown in Fig. 1, these radical reactions generate ozone form a chain reaction. VOCs are very important for photochemical ozone production in terms of the propagation of the chain reaction. Meanwhile, the role of NO_x in the ozone production is complicated. Whereas NO_x is the direct ozone precursor as shown in reactions (1), (1), (2), (2), it also destructs ozone as shown in reaction (3), (3). In addition, NO_x is an important terminator of the chain reaction to produce ozone due to the following reaction:

OH+NO₂+M→HNO₃+M.

In the case of low NO_x mixing ratios, ozone production rates rise when NO_x concentrations increase (NO_x-limited). Meanwhile, in the case of high NO_x mixing ratios, ozone production rates decrease when NO_x concentrations increase (VOC-limited) (e.g., Sillman et al., 1990; Milford et al., 1994). Hence, VOCs and NO_x form complicated interactions with ozone, so it is important to verify the relationship between ozone, VOCs, and NO_x in order to clarify the OWE in the urban atmosphere.

CARB (2001) proposes several hypotheses of the OWE, that is, “NO_x reduction,” “ozone quenching,” “NO_x-timing,” “carryover near the ground,” “carryover aloft,” “increased weekend emission” and “the aerosol and ultraviolet (UV) radiation hypothesis”. The first two hypotheses would be important to verify the complicated interactions between VOCs, NO_x and ozone for the OWE. The “ozone quenching hypothesis” is based on the titration of ozone by NO (reaction (3), (3)), which is emitted more abundantly in the urban atmosphere on weekdays than on the weekend. The “NO_x reduction hypothesis” is related to the chain reaction to produce ozone (Fig. 1). This paper reports on the current condition of the OWE in Japan and investigates its causes, focusing on NO_x and VOC chemistry. Tokyo and Osaka, the biggest cities in Eastern and Western Japan, respectively, were selected for this research. Fig. 2 shows the location of the Tokyo Metropolitan area and Osaka Prefecture in Japan.