A fbur-dimensional variational (4D-Var) data assimilation system for a regional dust model is applied to a heavy dust event which occurred between 30 March to 4 April 2007 over eastem Asia. The verdcal profiles of the dust extinction coefficients derived from 5 NIES LIDAR observation sites are directly assimilated. Themodeled dust extinction coefficients are improved considerably after assimilation. Assimilation results increase dust emissions over the Gobi Desert and Mongolia considerably; especially between 29 and 30 March. The heavy dust event is causged by the heavy dust uplift flux over the Gobi Deseit and Mongolia during those days. We obtain the total optimized dust emissions of 57.9Tg (57.8%laiger than before assimilation). We also confimed that the assimilated model results agreed with OMI aerosol index and WMO SYNOP report, and Lidar-4DVAR system has a potential to improve the accuracy of Asian dust emission/ transport model.

View full abstract

We have investigated the dynamics and formation mechanisms of the episodic pollution of photochemical ozone (O₃) during 8-9 May 2007 over Japan using hourly air quality data observed at monitoring stations in the whole Japan and a chemical transport model (CTM) in East Asia.
Observed O₃ concentrations at monitoring stations over Japan began to grow up in north Kyushu area around 900 JST on 8 May, and reached over 120ppbv in north Kyushu area, including isolated islands such as 116 and Gotou, and the west part of Chugoku area at 1500 JST. In night on 8 May, O₃ concentration fall down, but a part of stations in western part of Japan continued to keep the O₃ level over 120ppbv. On next day (9 May), O₃ concentrations are higher than those on previous day at 900 JST and then grew up over the wide area in Japanese Island excluding Hokkaido and the north part of Tohoku area. At 1500 JST, the high O₃ concentrations over 120ppbv were observed at lots of stations over the wide area in Japan: the surrounding areas of megacities (Tokyo, Nagoya, and Osaka) and Hokuriku (Toyarna and Niigata prefectures) and Setouchi areas.
CTM can reproduce the temporal variations of observed O₃ concentrations at monitoring stations during 8-9 May, but underestimates the peak O₃. This underestimation tends to be improved as increasing of emissions in north and central coastal areas of China. Modeled spatial distributions of surface O₃ during 7-9 May showed that a polluted air mass (O₃> 80ppbv) having a spatial scale over 500km was transported from the north coastal area of China to Japanese Island by the westerly strong wind at the northern part of moving high pressure system over the East China Sea. High concentrations of O₃ observed in Japan during 8-9 May were influenced by the transboundary air pollution of the photochemically produced O₃ by precursor gases emitted from China and Korea. Averaged contribution of Chinese emissions to surface O₃ during the high O₃ concentration (O₃> 80ppbv) in the same period is estimated to be over 25% throughout Japan excluding the northern part of Japan and to be 40-45% in Kyushu area. However, it should be noted that the model may fail to estimate the contribution of trasnboundary pollution in surrounding areas of megacities because the model doesn't have the capability to reproduce well the detail of urban air pollution. Finally, the modeled results and the observations of monitoring stations and Lidar shows that SO₂ gas and anthropogenic aerosols as well as O₃ are transported from Asian continent within a mixed layer below 1500m altitude.

View full abstract

Episodic pollution of photochemical ozone during 8-9 May 2007 over Japan was analysed using the nesting RAMS (Regional Atmospheric Modeling System)/ CMAQ (Community Multi-scale Air Quality) modeling system. The sensitivity based on emission scenario was calculated and the results of the scenario of appropriate ground were used for the analysis. In the simulation results, the features of this episode were reproduced well. The high ozone concentration area appeared on Sea of Japan side of the Kyushu and Chugoku region on May 8, and the area moved to east side of Japan and reached the Hokuriku region on May 9. In the model, the flow of highly polluted air masses from China to Kyushu and Chugoku region on May 8 was simulated, that suggested the influence of trans-boundary pollution from China on this episode. Actually, the influence from China was the largest, and the contribution rate was about 40% at a high concentration peak in North Kyushu and Chugoku region on May 8 and that in the east side of Japan was highest on May 9. Basically, our study concluded that this episode was strongly influenced by the trans-boundary pollution from China. However, the results were obtained by the scenario analysis approach and have uncertainty and there were some problems in reproducibility of the observation. Further research using the improved emission data and modeling system is needed.

View full abstract

Episodic high concentration of photochemical ozone exceeding the advisory-level (120 ppbv) during 8-9 May 2007 was analyzed using hourly mean concentrations of air pollutants in Japan.
On 8 May ozone increase started from Goto islands, which is located in a westernmost part of Japan. The daily maximum ozone in Oki was observed in the middle of the night. The peak time of ozone concentration lagged eastward along the coast of the Japan Sea. In remote islands, sulfur dioxide and particulate matter concentrations varied simultaneously with ozone. Backward trajectory analyses and vertical profiles of anthropogenic aerosols obtained by Lidar observation showed that polluted air-mass was transported from the Asian continent.
On 9 May high concentration of ozone was mainly found over eastern Japan. As in the preceding day, the time of daily maximum ozone lagged eastward along the coast of the Japan Sea. On the other hand, the time lagged northward to the interior of the Kanto Plain and ozone concentration was higher than that of the preceding day. Spatial and temporal variation of pollutants and meteorological conditions suggested that formation of land-sea breeze circulation accumulated urban air pollutants around the Tokyo metropolitan area.
These results showed that high concentration of ozone in Japan was caused by trans-boundary air pollutants originating from the Asian continent and domestic urban air pollutants enhanced the pollution level in the metropolitan area.

View full abstract

An “inverse analysis” method that provides the most probable location of the emission source and the emission rate corresponding to a series of measuredcon centration data is proposed. The METI-LIS (Ministry of Economy, Trade and Industry-Low rise Industrial Source dispersion) model is used for performing trial dispersion calculations at various source locations. Correlation coefficients between measured and calculated concentration data are plotted at the corresponding tentative source locations, and a contour map of the correlation coefficients iso btained.
Sakai area in Kinki idistrict, where high concentrations of ambient acrylonitrile have often been observed, was selected for the verification of the proposed method. By using the method with acrylonitrile concentrations measu. redata fixed station in the area during 2000-2004, a zolle where the probability of finding the source is high was determined. Additionally, short-termmonitoring of the concentration was performed at two points in 2006. By employing the abovementioned ilwerse analysis method and by considering this additional data, the location of the emission source was limited to a small area that overlapped with an acrylolnitrile-discharging plant reported in the PRTR (Poltutant Release and Transfer Register) system: the reported emission annount was only approximately one-forth or less of that estimated by the inverse analysis method. It should be noted, however, that this trial of the proposed method was rough because lt assunled a single emission source, and the nlethod should be developed further.

View full abstract

In this study emission amount of elemental carbon (EC) was estimated for the assessment of human health impacts in Kansai area. The emission inventory consists of two parts, one is emission from mobile sources and the other is stationary sources. Mobile sources were divided in that from roads, which is included not only trunk roads but branchy roads, and that from ships. Emission from stationary sources was reconstructed from the national emission database of soot and dusts. This emission inventory was developed mainly by the statistics in 2000. In Kansai area including 6 prefectures (Shiga, Kyoto, Osaka, Hyogo, Nara, and Wakayama), annual amount of EC emission was estimated at 6.62×10³ (t). According to the analysis of source contribution, emission intensity in holidays is higher than that in weekdays. And it was also shown that the emission from roads is occupied 92% in total EC emission, especially trucks are the major source of EC emission.
Moreover, with this inventory, some transport simulations in Kansai area by MM5 and CMAQ models were also conducted episodically in four seasons. Meteorological fields by MM5 had good performance in the comparisons with observed data. In summer simulation by CMAQ, comparably good agreement was shown in the comparison with time-averaged observed data, however, some underestimated points were found. In other seasons calculated values showed underestimated in the comparison with observed values, because air mass was mainly coming from main land of China and Korean peninsula. Therefore, future formation of modeling system, such in the determination of domains and the height of stationary sources, should be refined and other compounds would be included to be a integrated tool for the assessment of human health impacts.

View full abstract