60 AIR POLLUTION METEOROLOGY
mixing ability of the atmosphere is particularly
bad.
(3) Meteorological factors have to be taken into
account when evaluating air pollution control
measures. For example, the air quality in a region
many improve over a number of years—not as
a result of abatement measures, but because of
gradual changes in the weather characteristics. If
the effects of the meteorological changes are not
evaluated, efforts at abatement will be relaxed,
with the result of unsupportable conditions when
the weather patterns change again.Effects Between Source and ReceptorThe way in which the atmospheric characteristics affect the
concentration of air pollutants after they leave the source can
be divided conveniently into three parts:(1) The effect on the “effective” emission height.
(2) The effect on transport of the pollutants.
(3) The effect on the dispersion of the pollutants.Rise of EffluentTo begin with the problem of effluent rise, inversion layers
limit the height and cause the effluent to spread out hori-
zontally; in unstable air, the effluent theoretically keeps
on rising indefinitely—in practice, until a stable layer is
reached. Also, wind reduces smoke rise.
There exist at least 40 formulae which relate the rise of
the meteorological and nonmeteorological variables. Most
are determined by fitting equations to smoke rise mea-
surements. Because many such formulae are based only
on limited ranges of the variables, they are not generally
valid. Also, most of the formulae contain dimensional con-
stants suggesting that not all relevant variables have been
included properly.
For a concise summary of the most commonly used
equations, the reader is referred to a paper by Briggs
(1969). In this summary, Briggs also describes a series of
smoke rise formulae based on dimensional analysis. These
have the advantage of a more physical foundation than the
purely empirical formulae, and appear to fit a wide range
of observed smoke plumes. For example, in neutrally stable
air, the theory predicts that the rise should be proportional to
horizontal distance to the 2/3 power which is in good agree-
ment with observations. The use of dimensionally correct
formulae has increased significantly since 1970.
Given the height of effluent rise above a stack, an
“effective” source is assumed for calculation of transport
and dispersion. This effective source is taken to be slightly
upwind of a point straight above the stack, by an amount
of the excess rise calculated. If the efflux velocity is small,
the excess rise may actually be negative at certain wind
velocities (downwash).Transport of PollutantsPollutants travel with the wind. Hourly wind observations at
the ground are available at many places, particularly airports.
Unfortunately, such weather stations are normally several
hundred kilometers apart, and good wind data are lacking in
between. Further, wind information above 10 meters height
is even less plentiful, and pollutants travel with winds at
higher levels.
Because only the large-scale features of the wind pat-
terns are known, air pollution meteorologists have spent
considerable effort in studying the wind patterns between
weather stations. The branch of meteorology dealing with
this scale—the scale of several km to 100 km—is known as
mesometeorology. The wind patterns on this scale can be
quite complex, and are strongly influenced by surface char-
acteristics. Thus, for instance, hills, mountains, lakes, large
rivers, and cities cause characteristic wind patterns, both in
the vertical and horizontal. Many vary in time, for example,
from day to night. One of the important problems for the
air pollution meteorologist is to infer the local wind pattern
on the mesoscale from ordinary airport observations. Such
influences are aided by theories of sea breezes, mountain-
valley flow, etc.
In many areas, local wind studies have been made.
A particularly useful tool is the tetroon, a tetrahedral bal-
loon which drifts horizontally and is followed by radar. In
some important cities such as New York and Chicago, the
local wind features are well-known. In general, however, the
wind patterns on the mesoscale are understood qualitatively,
but not completely quantitatively. Much mesoscale numerical
modeling is in progress or has been completed.Atmospheric DispersionDispersion of a contaminant in the atmosphere essentially
depends on two factors: on the mean wind speed, and on the
characteristics of atmospheric “turbulence.” To see the effect
of wind speed, consider a stack which emits one puff per
second. If the wind speed is 10 m/sec, the puffs will be 10 m
apart; if it is 5 m/sec, the distance is 5 m. Hence, the greater
the wind speed, the smaller the concentration.
Atmospheric “turbulence” consists of horizontal and
vertical eddies which are able to mix the contaminated air
with clean air surrounding it; hence, turbulence decreases the
concentration of contaminants in the plume, and increases
the concentration outside. The stronger the turbulence, the
more the pollutants are dispersed.
There are two mechanisms by which “eddies” are formed
in the atmosphere: heating from below and wind shear.
Heating produces convection. Convection occurs when-
ever the temperature decreases rapidly with height—that is,
whenever the lapse rate exceeds 1C/100 m. It often pen-
etrates into regions where the lapse rate is less. In general,
convection occurs from the ground up to about a thousand
meters elevation on clear days and in cumulus-type clouds.
The other type of turbulence, mechanical turbulence,
occurs when the wind changes with height. Because thereC001_005_r03.indd 60C001_005_r03.indd 60 11/18/2005 10:13:36 AM11/18/2005 10:13:36 AM