To determine reasonable emission reduction measures for the mitigation of photochemical oxidant (ozone) pollutions over the Kanto area, it is important to know whether the sensitivity of the ozone concentration to precursor emissions is VOC-sensitive or NO_x-sensitive at each location. Using a three-dimensional chemical transport model (CTM), we estimated the spatial distribution of the sensitivity regimes for the daily maximum ozone concentrations at each location over the Kanto area for a total of three days on which warning-level ozone was formed within the Kanto area. The Biogenic VOC (BVOC) emission amounts, input to the CTM, were estimated by considering recent emission measurements for Japanese native plant species. The results showed the sensitivity regime was VOC-sensitive over the limited regions near or within the urban center and was NO_x-sensitive over most regions including the areas which had many warning-level ozone concentration days. On the other hand, the simulation results obtained using previous lower BVOC emission estimates showed that the regions with VOC-sensitive regimes significantly increased, indicating that the BVOC emission amounts input to the CTM significantly affects the estimated ozone concentration sensitivity regimes.

View full abstract

It has been proposed that the sensitivity regimes of ozone concentrations to precursor emissions can be estimated by measurements of the parameters called photochemical indicators, i.e., the concentration ratios of several secondary formed species, for each location. In the present study, we first evaluated whether these parameters were also applicable for the Kanto area, Japan, using the simulation results from a companion paper. This evaluation indicated that the ozone / total nitrate concentration ratio was applicable with the derived threshold values of 8.6 (or below) and 9.0 (or above) for discerning the VOC-sensitive and NO_x-sensitive regimes, respectively. Hence, we measured the total nitrate concentrations and derived the ozone / total nitrate concentration ratios at five sites in Tokyo and Saitama prefecture on several days with a clear sky and high air temperature in the summer, during which warning-level ozone was easily formed, and then evaluated the sensitivity regimes at these locations. The results showed that the sensitivity regimes were NO_x-sensitive at all the measurement sites except for two sites within the urban center. These results, based on the measurements of the ozone / total nitrate concentration ratio, were almost consistent with the simulation results from the companion paper, obtained using BVOC emission estimates which were derived by considering recent emission measurements for Japanese native plant species.

View full abstract

Real-time monitoring of individual pollutants in the exhaust gas of a mode-driving diesel truck was conducted utilizing the laser multi-photon ionization technique. The volume mixing ratio of benzene, toluene, o-xylene, m-xylene, p-xylene, and phenol were individually analyzed. Temporal variations in these species during the driving mode were successfully observed every second. From the observed mixing ratio and flow rate of the exhaust, the emission factor and OH reactivity of each compound were acquired. As a result, in some cases, unique trends were observed, which could not be captured by monitoring the total hydrocarbons (THC) and/or the mode average of each compound. Especially, a significant phenol level was observed during the high-speed driving period. Additionally, the time series data with a 1-sec resolution were categorized by the driving speed of the vehicle. Consequently, it was found that the emission factor and OH reactivity per travel distance reached the minimum in the middle speed range (50-70 km/h) for the target VOCs. It was confirmed that the real-time analysis of individual compounds in the exhaust gas during the driving mode would be useful.

View full abstract

The inter-comparison of chemical transport models (CTMs) of four research groups was conducted for O₃ and PM_2.5 simulations in the Kanto area in summer 2007. Three groups used CMAQ v4.6 and the other employed CMAQ v4.5; each group used different emissions data. The results clearly showed that all the CTMs reproduced well the temporal variation of O₃ (r > 0.5 at all four sites), but all the CTMs underestimated PM_2.5 concentrations and did not reproduce the PM_2.5 temporal variation (r <0.5) at suburban sites. Among the PM_2.5 species, the concentration and temporal variation of SO₄2^- was predicted well by all the CTMs (within a factor of 2 and r > 0.5 in most cases). By contrast, the elemental carbon (EC) and organic aerosol (OA) concentrations were underestimated by all the CTMs. The predicted NO₃^- concentrations showed a high variability (up to a factor of 10) among the four CTMs. Differences in the predicted concentrations of the primary products were explained well by the differences in their emission rates. However, differences in the secondary products, including O₃, TNO₃ (= HNO₃+ NO₃^-), and the secondary organic aerosols, were not explained by the differences in the precursors' emission rates or in boundary conditions.

View full abstract

To analyze the contribution of Asian-originated air pollutants, year-round monitorings of PM_2.5 have been conducted on Fukue Island and in Fukuoka City in Kyushu, the western part of Japan. In April, 2009, hourly PM_2.5 concentrations in Fukuoka and Fukue fluctuated synchronously, with maximum concentrations of 90 and 80 μgm^-3, respectively, but with about 12 hour delay in Fukuoka. In addition, the PM_2.5 concentrations on Fukue were the same as, or slightly higher than, those in Fukuoka during the high concentration events. The further analyses of the aerosols composition collected during the intensive measurement period in April showed that sulfate particles accounted for the largest portion of the high-concentration PM_2.5, followed by particulate organics. During the period of the prominent high-concentration events, the non-seasalt sulfate concentrations were higher on Fukue than in Fukuoka. The PM_2.5 concentration on Jeju Island, Korea, located west of Fukue Island started to increase before that of Fukue Island, which indicated these high-concentration events at Fukue and Fukuoka are due to long-range transport rather than domestic air pollution. The surface weather chart during the events were those of typical long-range transport patterns, e.g.,“ behind the front” and “circular flow around migrating anticyclones”. The monthly average PM_2.5 concentration on Fukue Island in April was also higher than that in Fukuoka City. These results indicate that in northern Kyushu, even in a large city, the PM_2.5 concentration during the spring of 2009 was subject to regional pollution rather than domestic air pollution.

View full abstract

The present study developed regression equations to predict minute ventilation rate (V_E) of young children using a tri-axial accelerometer. Body acceleration during 9 different activities was measured on 28 Japanese preschool children using ActivTracer, and then calibrated against V_{E, BTPS} measured by the Douglass bag method. We observed a significant correlation between the V_{E BTPS} per kg body weight and the synthetic accelerometer measurements across all activities (R² = 0.925, n = 222, p <0.001). However, the linear regression equation overestimated the V_{E, BTPS} per kg body weight for the resting state, playing with plastic bricks, and walking by 12-18%, while it underestimated the value for ball tossing and stair climbing by 18-22%. The agreement between the estimation and observation was significantly improved by discriminating the activity within a specific range of accelerometer measurements, between ‘walking’ or ‘other than walking’, based on the vertical and horizontal body accelerations, and correction of the regression equation for the latter activities. Approaches based on the energy consumption rate used in previous studies underestimated or overestimated our V_E for resting state measured by the Douglass bag method by -19% to +33%, while our regression model using measurements from the ActivTracer by -6 to +16%, indicating our regression equations better estimate the V_E of daily life activities in children.

View full abstract