996 
be lower at times and higher at other times than the 
air above; deviations in humidity, however, are almost 
always in the same direction. Just as wind velocity is 
lower near the earth’s surface, so the absolute and rela- 
tive humidities are always higher. Only in the morning 
hours is the ground surface sufficiently colder than 
the air for water vapor, in the form of heavy dew, to 
condense there in a quantity sufficient to dry out 
temporarily the lowest meter of the atmosphere. (Often 
this applies only to a layer several centimeters deep.) 
A relative humidity at 10 em which is 30 to 50 per cent 
higher than that at 2 cm is not rare. Thus the climate 
near the ground is also entirely different in this respect. 
Rapid fluctuations of humidity with time are as 
characteristic of the microclimate near ground as are 
temperature fluctuations. Air parcels enriched in mois- 
ture content rise upward from the ground, dry parcels 
sink downward. Because of poor mixing they cause the 
indicators of special hygrometers to move back and 
forth in Jumps. When the ground is covered with 
drought cracks, saturated ‘air rises from the depths of 
these crevices and causes deep layers to dry out more 
quickly. 
Moisture layers near the ground are sometimes vis- 
ible in the form of flat fog banks in enclosed spaces or 
as the steam above highways which is so annoying to 
the motorist (for imstance after a shower strikes a hot 
road surface). However, the formation and dissolu- 
tion of fog above the ground, as well as the formation 
of dew, have not yet been investigated sufficiently. 
Exact investigation is rendered difficult by the dis- 
continuous structure of the layer near the ground, 
varying conditions of exchange, and the effects of col- 
loido-chemical and aero-electrical processes superim- 
posed on the thermodynamic changes. Moreover, 
humidity changes in the ground air as well as conden- 
sation within the ground must also be taken into con- 
sideration. Investigations along these lines seem prom- 
ising and may provide new information basic to the 
entire field of microclimatology. 
Wind Conditions. Today we are well informed on 
the variations of the wind speed in the air layer near 
the ground. All measurements in the air layer near 
ground and water surfaces are, according to investiga- 
tions by Thornthwaite and Halstead [32], represented 
with sufficient accuracy by the equation 
uP = (log z — log 29)/log a, 
where wu is the wind speed (m sec~) at the height z 
(m); p is a parameter whose value depends on the 
magnitude of the vertical temperature gradient. If 
p = 1, the equation above is transformed into the 
logarithmic Jaw that was formerly in general use. In 
this case, the observed values, when graphically shown 
in a coordinate system with u as abscissa and log z as 
ordinate, lie on a straight line whose slope is deter- 
mined by the value of log a and whose point of inter- 
section with the ordinate is determined by the mag- 
nitude of log zo. 
However, in the same coordinate system, the meas- 
ured values do not lie on a straight Ime but on a curve 
CLIMATOLOGY 
that is concave upward, if the temperature stratifica- 
tion is unstable. In reverse, during a stable nocturnal 
inversion a curve is found that is convex upward. The 
parameter p in the equation above takes care of this 
influence of the vertical temperature gradient. When 
a strong temperature decrease exists with increasing 
height, then p > 1; with great instability, » may reach 
the value of 2. For stable stratification p < 1. For 
the smallest amount of turbulent mixing of the air 
that may occur in practice, p has the value of approxi- 
mately 0.5. 
According to the equation given above, a value 
uw = 0 is reached at a height 2) . The value of 2) (rough- 
ness height) depends on the nature of the surface and 
on the existing meteorological conditions. Over water, 
snow, and level bare soil, a value of z) = 0.02 m can 
be used as an approximation. 
Eddy Diffusion. Whereas the variation of wind with 
height above the ground is very well known, the de- 
termination of the austausch coefficient A represents 
one of the most difficult and immediate problems of 
microclimatology. 
At the surface, a boundary layer of about 1-mm thick- 
ness is assumed, in which molecular processes (molecu- 
lar heat conduction, and diffusion) are almost exclu- 
sively effective. Above this layer, eddy diffusion deter- 
mines the vertical exchange of all properties in the 
air layer near the ground. Although the value of A is 
of greatest importance for the understanding of the 
total heat and water balance at the ground, only a few 
measurements of A in the air space near the ground 
are available. 
The measurement of A is accomplished by the simul- 
taneous observation of the short-period fluctuations 
of a meteorological element (temperature, momentum, 
water-vapor content) and its vertical gradient. This 
method assumes that, for example, with a very unstable 
temperature stratification a relatively warm air parti- 
cle comes from below, a relatively cold one from above. 
Since the same mixing process exchanges all proper- 
ties simultaneously, it might be expected that, at a 
given place and time, the same value of A would be 
found independently of the meteorological element that 
is employed for the determination of the magnitude of 
A. This universal character of the austausch forms the 
basis of many computations of the heat balance of our 
earth. However, more recent measurements show that 
quite different values of A are obtained when different 
elements are used. Thus, the austausch proceeds in a 
different manner for the different elements (unless the 
differences are attributed to the properties of the in- 
strumental arrangement). None of the several explana- 
tions found in the literature is completely satisfactory. 
Frankenberger [8] made simultaneous measurements 
of the microstructure of the horizontal and vertical 
wind speed by means of a very sensitive instrument 
and showed that there exist two entirely different 
types of austausch motion. As regards the magnitude 
of this motion, he proved not only that faster air 
particles come from above, slower ones from below 
(which represents vertical motion promoting aus- 
