In recent years, high-order turbulent statistics such as the second and higher moments of wind fluctuations, turbulent dissipation rate, and time scale have been often used to predict concentration distributions. Thispaper describes atmospheric dispersion models using high-order turbulent statistics, with special focus on the Lagrangian particle dispersion models. The Lagrangian particle dispersion models have the following features:
(1) The models are applicable to the analyses of diffusion phenomena in complicated flow fields generated by topography, heat and plants.
(2) The concept of the Lagrangian method is natural and easily understandable. Unlike the Eulerian model, the procedure of numerical solution is almost free of problems.
(3) A relatively small amount of work is required to establish the Lagrangian particle dispersion models compared with closure models.
Lagrangian particle dispersion models have been studied for nearly twenty-five years. During this period, the models have been developed to take consideration of the inhomogeneous and non-Gaussian turbulence. These models have been used successfully in the prediction of plume in the convective boundary layer. The models are expected to be useful in the prediction of the three-dimensional diffusion in a complex terrain, the diffusion analysis within a plant canopy or an urban canopy, and the prediction of concentration fluctuations

View full abstract

Two psychophysical experiments were made to explore the attribute of “odor” and “pungency” in respect of inhalation time and continuous exposure. Four or three subjects judged perceived intensity of odor and pungency in them. In the first experiment perceived odor intensity increased gradually according as the inhalation time became longer at lower concentration level, but decreased at higher. On the other hand perceived pungency increased extremely at any condition, and the importance of the whole quantity of substances represented as the product of inhalation time and concentration was revealed. These findings suggest the necessity to keep inhalation time of each subject constant on odor bag test. In the second, continuous exposure of odorants caused perceived odor intensity to decrease exponentially, but caused perceived pungency to increase and fluctuate. These results represent that much attention must be paid to the attribute of pungency on estimating continuous exposure of odorants.

View full abstract

Thirty-six kinds of pesticides in air were monitored at two urban areas and one residential area located near the agricultural area in Kitakyushu.
Sampling was conducted in four days periods in four months (April, August, November and February) in 1992-1993.
Thirty-five species of pesticides were detected in air. Their concentration range was 0.01-70ng/ m³. Although the concentration level of 30 species of the detected pesticides was less than 1ng/ m³. The concentration of pesticides in the air sample in August was the highest. This fact was clear in the sampling site near the agricultural area. Especially six kinds of pesticides, including Flutoluanil and Fenobucarb were found at high concentration. These pesticides are used not only agricultural areas but also on golf courses or other places. From these results, it was clear that the concentration of pesticides in air is influenced by many kinds of emission sources. The concentrations of pesticides in air in the other three months were relatively low and the fluctuation of the concentration level was small. The total amount of pesticides detected on the XAD-2 resin trap was higher than that on the quartz fiber filter in three months except August. However, in August, a high concentration of Flutoluanil was found on the filter. It was suggested that a large amount of Flutoluanil was used in the agricultural area just before the sampling time.

View full abstract

High concentrations o ozone, principally stratospheric origin, are observed in spring at the northern Kyushu district. In order to estimate the long-term ozone variation and the effects of the weather condition, the observations of ozone were made for about 6 years at an unpolluted mountain-peak, and the concentrations of beryllium-7 were measured for 50 days in spring and fall.
The monthly maximum ozone at the mountain-peak was mainly observed in April and May, having the values of 44-60 ppb as the averages of all l-hr ozone during the months. The seasonal variation was almost similar to that in the northern hemisphere, where ozone maxima in spring and ozone minima in fall have been observed. Ozone concentrations at the mountain-peak were slightly reduced year by year, and the rate of this reduction was estimated as 4.7%/ Year.
Average concentrations of beryllium-7 were (5.7±2.0)×10o³ Bq/ m³ in spring and (5.1±2.0)×10^-3 Bq/ m³ in fall, Concentrations of beryllium-7 were increased under these meteorological conditions such as the behind of a cold front and the behind of high-pressure. The ozone concentrations at the mountain-peak are relatively high under these meteorological conditions such as a troughof atmospheric pressure and atmospheric high-pressure system. The ozone concentrations affected by a northerly air mass at mountain-peak are higher than those by a southerly air mass.

View full abstract

Simulation of SPM concentration was carried out with taking into account the effect of water, since the weight of SPM measured in an urban area appeared to be affected by liquid water through change of relative humidity of the air. As a diffusion model, so-called multi-box model was applied. Both of pseudo-diffusion arose from calculation scheme of the multi-box model and actual turbulent diffusion affect on the concentration. Thusthe correction of the pseudo-diffusion was discussed first, and a method to correct this effect was proposed. This correction method leads to reflecting actual turbulent diffusion, resultantly.
Effect of liquid water on the weight of particles in the air was estimated by simplified Winkler's⁸ empirical formula introduced for the description of characteristics of aerosol in an urban area. Calculation was made for 4 days, chosen from high concentration days in winter in Tokyo prefecture. With taking into account the effect of water, calculated concentration became closer to observed data in comparison with no water condition.

View full abstract

In the Kanto plain, severe air pollution due to suspended particulate matter (SPM), nitrogen oxides (NO₂), etc. occur most frequently in early winter. The meteorological structure of calm and stable condition has been analyzed statistically in connection with the high-level pollution in the three Decembers from 1989 to 1991.
The daily average concentration of SPM has a high correlation with that of NO₂, and is very uniformly distributed over a wide region including the inland portion of the Tokyo metropolitan area and the central part of Saitama Prefecture.
On most high-SPM days (daily average ≥130 μg/m³), hourly concentration does not lower below 100 μg/m³since the previous nights, suggesting that weather systems with a time scale longer than one day cause the very high concentrations.
On those days, the temperature difference between Mt. Tsukuba (870m ASL) and the Kanto plain surface below 100 m ASL almost vanishes in the daily average. This results from synoptic-scale warming in the upper layer.
The differencial cooling compared with the low-SPM days (daily average <90 μg/m³) is larger by about 2 K in the central part of the plain than in the areas along the mountains (above 100m ASL). It can be adequately called as a cold air lake. In this region, a very calm condition lasts because no local wind system affectsthere, while the sea breeze develops along the shore of Tokyo Bay and the valley winds along the mountains:

View full abstract

[in Japanese]

View full abstract