ATMOSPHERIC POLLUTION 
a result of subsidence occurring over the eastern portion 
of the Pacific anticyclone. Near the surface the air is 
relatively cool because of its recent oceanic trajectory, 
and mechanical turbulence has produced in it a marked 
lapse rate. Between these two bodies of air an inversion 
is usually present, extending perhaps from 2000 to 4000 
ft. Thus, although there may be turbulence and convec- 
tion near the surface, the inversion aloft mhibits the 
upward diffusion of contaminants by eddy processes, 
and extensive pollution develops. 
Effect of Rain. It is not yet clear whether rain tends 
to alleviate surface pollution by washing impurities 
out of the air. At Leicester it was found that from 27 to 
37 per cent of dissolved matter deposited on the surface 
and from 16 to 19 per cent of insoluble matter deposited 
there was brought down by rain. It is not known, how- 
ever, whether these materials were present in the cloud 
particles from which the rain developed or whether 
they were washed out of the contaminated atmosphere 
by the rain drops as the latter fell. An investigation in 
fourteen large American cities [42] indicated that rain 
had no effect in cleansing the air, but rather that it 
tended to increase the concentration of smoke near the 
ground. The evidence at Leicester indicates that rain 
does not reduce the concentrations of sulfur dioxide in 
the air at all rapidly. 
The Annual Variation. Near cities in extratropical 
latitudes there is a marked annual variation, with a 
maximum in winter and a minimum in summer. A 
large proportion of the winter pollution is due to prod- 
ucts of combustion released from heating plants in 
buildings, but industrial contaminants emitted are not 
subject to the same degree of annual variation. Be- 
cause of this complication, the effects of the annual 
variation of atmospheric conditions on pollution are 
not at all well known. The winter maximum and sum- 
mer minimum are well marked at New York City [17] 
and at Leicester. At the latter, smoke has the largest 
range of annual variation, and insoluble deposit the 
smallest. 
The Weekly Variation. This is due, not to mete- 
orological factors, but to the shutting down, partially 
or completely, of industrial plants durmg the week 
end. It will therefore not be discussed here. 
The Daily Variation. Surveys of smoke and dust in 
American cities, such as those for New York [17] and 
for a group of fourteen large cities [42], and in British 
cities, such as London, Glasgow, Blackburn, and Stoke- 
on-Trent [28], show a maximum at times ranging from 
6:30 to 9 a.m. and a secondary maximum in the late 
afternoon. Increased industrial activity and home and 
office heating in the morning taken in conjunction with 
increasing turbulence associated with solar heating, as 
outlined earlier, account for the morning maximum; 
decreasing turbulence in the late afternoon before in- 
dustrial emission of pollution has been reduced causes 
the secondary maximum. 
Effect of Topography. As with single sources, the 
pollution from multiple sources, as represented by an 
industrial city, is markedly affected by topographic 
features. In a valley, for example, under stagnant 
1147 
atmospheric conditions, serious concentrations may and 
sometimes do develop; the valley sides act as physical 
barriers which prevent the lateral diffusion which nor- 
mally occurs over more level terrain. Two extreme 
examples may be quoted. During the period December 
1-5, 1930, severe pollution occurred in the Meuse valley 
in Belgium [22]. Anticyclonic conditions prevailed dur- 
ing the period, with fog and very light winds which 
carried pollution from the city of Liége and from 
industrial plants nearby into a narrow portion of the 
valley; an inversion accentuated the pollution. The 
conditions were so severe that several hundred persons 
suffered from acute respiratory troubles and sixty- 
three died on December 4 and 5. A similar disaster 
occurred at Donora, Pa., during the last few days of 
October, 1948 [23, 99]. Donora lies in the relatively 
deep and narrow valley of the Monongahela River at a 
distance of about twenty miles from Pittsburgh. Dur- 
ing the five-day period of light winds associated with 
anticyclonic conditions, the pollution became severe 
and twenty persons died and hundreds were stricken 
in and near Donora. It appears from these instances 
that if stagnant anticyclonic conditions persist for four 
days or more, an occurrence most probable in middle 
latitudes in the autumn, fatal concentrations of pollu- 
tion may develop. Topographic effects on pollution 
near some great cities are important. For example, the 
mountain ranges near Los Angeles are a factor at certam 
times in increasing the pollution problem there [54]. 
Because of the proven danger of lethal concentrations 
of contaminants to valley communities, there is an 
urgent need for both theoretical and observational 
programs of study of valley pollution. Given a knowl- 
edge of certain meteorological quantities in and just 
above a valley, of the rate at which contaminants are 
entering the valley, and of the topography of the 
valley, it should be possible to compute from approxi- 
mate expressions the rate of accumulation of impurities 
in a given section of the valley during prolonged stag- 
nant atmospheric conditions. Such data would fore- 
warn of approaching critical concentrations and of the 
necessity for preventive action. As outlined in a later 
section, the climatological use of such data would also 
be most valuable. 
Serious accumulations of contamimants are more 
likely to occur at inland locations where, in general, 
surface winds are lighter than near the coast line [104]. 
Further Research. There is a need for many more 
measurements of pollution in, near, and above industrial 
cities and of the associated pertinent meteorological 
parameters. Relatively little attention has been paid 
to concentrations at large distances from industrial 
cities; in the Leicester report [24, p. 126] it is suggested 
that downwind from a city the size of Leicester the 
smoke diminishes as the distance for the interval 4 to 
10 miles; farther from the city, from say 10 to 100 
miles, it diminishes as the square of the distance; and 
beyond about 100 miles it diminishes more slowly 
again, and ultimately as the distance. Such a tentative 
formulation of large-scale aspects is useful, but observa- 
tions are needed to permit confirmation or modification 
