Size distributions of total suspended particulate matter (TSP) and various elements were predicted by chemical element balance method and compare with observed ones. The size distribution of particles originating from main source components were estimated separately. The predicted size distribution of TSP agreed fairly well with the observed one in coarse particle region, but disagreed appreciably in fine particle region. The predicted and observed elemental size distributions agreed with each other in a wide range of particle size for non-volatile, lithophile elements, but disagreed appreciably for the elements, S, Cl, Cr, Co, Ni, As, Se, Cd, Sb, I and W. Possible explanations for these disagreements were given. The comparison of predicted and observed size distributions, in addition to the comparison of predicted and observed atmospheric concentrations, will be useful for deepening the understanding about the contributions of various aerosol source components and for testing the quality of the prediction. The following items are regarded to be important for improving the chemical element balance method attempted in the present study.(1) Appropriate elemental compositions of particles originating from iron-and-steel industry and diesel engine exhaust should be determined.(2) Accurate data on the secondary aerosol formation from SO₂, NO_x and hydrocarbons should be collected.(3) A new method to make better prediction for volatile elements should be developed, in which it is taken into consideration that they condense on particles upon cooling in ambient air after escaping from high temperature flues.

View full abstract

The pattern of daily change of death number in Osaka City area comparing with that in control area during the period of 1962-1972 was observed. The high level of death number (Excess death) was selected by the deviation from the standard line. The standard line was set by two methods. One is the line showing the seasonal trend by the 15 days moving average and another one is the trend line corresponding to the temperature, which is shown by the regression curve between death number and temperature.
The air pollution effect was distinguished from combined atmospheric effect through the comparison with control area. It was shown that the excess death appeared often in polluted area according to the increase of air pollution level.

View full abstract

The type of excess death pattem consists of 3 types, which are weather type, air pollution type and combined type. Each type was analyzed through the pattern of relationship between changes of atmospheric conditions and daily death number with the lapse of time.
The characteristic difference among these patterns is not clear and it tends to be that the effect of air pollution and temperature to heart disease is immediate and that to respiratory disease or to old persons occurs with time lag.
In the analysis of death number change during the period of smog episode in December 1970, the increase of death by chronic rheumatic heart disease in that period was remarkable.

View full abstract

The effect of air pollution and temperature to death number is usually appeared as a complex. The death number change was analyzed on a certain temperature zone by air pollution in order to remove the influence of temperature.
The concentration of each pollutant was revised to pollution index based on the ratio to the threshold stimulus factor and each index level was summed to yield the combined index level. It was suggested that the doseresponse relationship between death number and pollutant-combined index was formed.
Concerning the index of temperature, it was devised based on the deviation from the temperature zone of 10°C-26°C and was summed to yield the combined index level with pollution index. The death increase caused by chronic rheumatic heart disease is remarkably connected with the change of atmospheric conditions.

View full abstract

A study was made of relationships between the sampling interval and the estimated accuracy of such statistical values of population as annual mean value and percentile values for one year, both of which are obtained from sampled statistical values in the case of sampling made of air pollution at an equal-time interval.
By the application of the stratified sampling theory to this method of sampling, relationships between the estimated standard errors of some percentile values for the log-normal distribution and the number of samples are calculated and are shown in diagrams, whereby the results were compared with the actual data of errors.
It follows from the results that the calculated standard errors of the percentile values of the sample data shown in the diagrams were slightly overestimated, as compared with the actual data of errotr.
It is suggested, however, that these diagrams may be used in choosing the sampling time duration in the intermittent sampling procedure at an equal-time interval.

View full abstract

Exhaust gas from a wet-oxidation treatment plant for night soil was analysed by GC-MS-SIM, GC-ECD, GC-FPD, GC-FID, etc. Thirty-three components could be determined quantitatively. Among them, the following compounds are worthy of attention because of their concentrations: furan (70 ppm), dichlorobenzene (84 ppm), formaldehyde (260 ppm) and acetaldehyde (40ppm).

View full abstract