It is difficult to widely cover air pollution in roadside areas with a limited number of air pollution monitoring stations. Therefore, we focused on the biomonitoring technique using leaves from the roadside trees. In this study, Ginkgo leaves sampled along major arterial roads in Osaka city in the spring, summer and autumn of 2014, 2015 and 2016 were used as the biomonitor. The chemical composition of the particles retained on the leaves was quantified by PIXE (Particle-induced X-ray emission) analysis. The total mass of the particles on the leaves increased with the time between the foliation and sampling date. The particles on the leaves were affected by traffic related sources, with a greater influence by yellow sand in the spring than in the other seasons. The elemental ratios of road dust, brake wear dust and automobile exhaust gas were similar to that of the particles on the leaves. This result suggested that the biomonitoring method in this study captured the characteristics of the roadside air pollution. The elemental ratio obtained by the biomonitoring had different characteristics from that of the SPM (Suspended Particulate Matter) measured at the neighboring roadside monitoring station, because particles larger than 10 μm in diameter can be captured by the leaves and water soluble particles on the leaves can be washed away by the rain. The biomonitoring result using Zelkova leaves shared common characteristics with that of Ginkgo leaves, indicating that the biomonitoring technique can be applied to other species of plants. These results showed that the biomonitoring technique is a potentially effective method to assess roadside air pollution.

The Community Multiscale Air Quality Model (CMAQ) was used to estimate the contributions of long-range transport (LRT) from the Asian Continent to Japan, domestic emissions in Japan, and background sources including boundary inflow and natural emissions to the concentrations of nitrate aerosols and its precursor compounds (NH₃, HNO₃, NO₂) in four major Japanese urban areas in fiscal year 2010. The estimated LRT, domestic and background contributions to the PM_2.5 concentration in Japan were 32–54%, 18–37% and 28–31%, respectively. The estimated LRT contribution to the PM_2.5 was generally higher in the western areas of Japan (closer to the continent). The contribution of the LRT was classified into the direct LRT contribution and indirect LRT one representing the PM_2.5 produced from the domestic and transported precursor compounds. The indirect LRT contributions in each area were dominated by NH₄NO₃ and had no significant difference in the rate and the total amount of chemical components. Moreover, the formation of domestic nitrate aerosol was affected by the domestic NH₃ and overseas HNO₃, according to the results of source contributions of precursor compounds. The indirect LRT contribution can be reduced by controlling the NH₃ emissions in Japan.

The daily concentrations of the PM_2.5 components were measured from fiscal year (FY) '04 to '14 at a suburban roadside site in Nagakute. The annual means of the PM_2.5 mass concentration tended to decrease during the observation period, and the decreasing rate became lower after FY '10. The OC and SO₄^2- indicated higher annual mean concentrations than the other components of PM_2.5, suggesting the importance in the decrease of both components for improvement of the annual mean PM_2.5 concentration. The annual mean concentration of the EC showed a downward trend from FY '04 to FY '09, and it is considered that the substitution effect of diesel cars to newly regulated vehicles is the main factor. On the other hand, from FY '09 to FY '14, the EC moderately declined, and the substitution effect became lower. The annual mean concentrations of the OC, which were higher than the EC, showed a temporal variation similar to the EC from FY '04 to FY '07. However, from FY '07 to FY '10, the mean concentrations of the OC rarely varied in spite of the decrease in the mean concentrations of the EC. The means of the main component (EC, OC, SO₄^2-, NO₃^-, NH₄⁺) concentrations of 5% days for the PM_2.5 ranking of the highest, medium and lowest were evaluated for each FY. The ratio of SO₄^2- of the highest rank was higher than the other ranks in each FY, which revealed the importance of measures to lower the SO₄^2- concentrations for decreasing the 98 percentile value from the lowest daily concentration of PM_2.5.