L
X
FIG. 1
that this scattering phenomenon depends on the wind structure, on the
frequency and intensity of the gusts, has been recognized for a long time; if
one observcs the same chimney in the early hours, before sunrise, when the
wind blows slowly and smoothly, thc smoke looks like a narrow ribbon,
propagating very quietly, sometimes to very great distances, before disappearing in thc atmosphere. The diffusion of smoke by wind gusts is a typical
example of turbulent atmospheric diffusion; the word turbulent specifies
clearly that the main cause ofthe diffusion is attributed to the fluctuations of
the wind velocity and that molecular diffusion, ever-present in any fluid
motion. contributes a negligible part to the observed phenomenon.
Turbulent diffusion is clearly a particular aspect of the large class of facts
descrihd under the general heading of turbulence; but it must be pointed
o u t that during the development that took place in the early history of this
field, both theoretical as well as experimental progress of our knowledge on
turbulent diffusion have played a foremost role in the developmcnt of our
ideas about turbulence. For instance, it was from his observations of the
diffusion of a colored streamline in a pipe that Osborne Reynolds (1883)
made thc fundamental distinction between two types of flow in a pipe and
introduced the " Reynolds number." the nondimensional parameter which
11ab pioved to bc of paramount importance; the pioneering work of Sir
of NCH Foundland.
3.
Among the different types of turbulent diffusion the case of atmospheric
diffusion is spccially difficult, because a well-adapted theoretical frame has
been lacking during the period we have to consider here; the theory of
turbulence WBS first developed in the case of a homogeneous (even most of
X
FIG. 1
that this scattering phenomenon depends on the wind structure, on the
frequency and intensity of the gusts, has been recognized for a long time; if
one observcs the same chimney in the early hours, before sunrise, when the
wind blows slowly and smoothly, thc smoke looks like a narrow ribbon,
propagating very quietly, sometimes to very great distances, before disappearing in thc atmosphere. The diffusion of smoke by wind gusts is a typical
example of turbulent atmospheric diffusion; the word turbulent specifies
clearly that the main cause ofthe diffusion is attributed to the fluctuations of
the wind velocity and that molecular diffusion, ever-present in any fluid
motion. contributes a negligible part to the observed phenomenon.
Turbulent diffusion is clearly a particular aspect of the large class of facts
descrihd under the general heading of turbulence; but it must be pointed
o u t that during the development that took place in the early history of this
field, both theoretical as well as experimental progress of our knowledge on
turbulent diffusion have played a foremost role in the developmcnt of our
ideas about turbulence. For instance, it was from his observations of the
diffusion of a colored streamline in a pipe that Osborne Reynolds (1883)
made thc fundamental distinction between two types of flow in a pipe and
introduced the " Reynolds number." the nondimensional parameter which
11ab pioved to bc of paramount importance; the pioneering work of Sir
of NCH Foundland.
3.
Among the different types of turbulent diffusion the case of atmospheric
diffusion is spccially difficult, because a well-adapted theoretical frame has
been lacking during the period we have to consider here; the theory of
turbulence WBS first developed in the case of a homogeneous (even most of