Geoscience Reference
In-Depth Information
Accordingly, Slatyer and McIlroy (1961) reasoned that the first term on the right of
Equation (4.23) may be considered a lower limit for evaporation from moist surfaces.
Thus
+
γ
E
e
=
Q
ne
(4.30)
was referred to as equilibrium evaporation, and the second term of (4.23) may be inter-
preted a departure from that equilibrium. In the absence of cloud condensation or radia-
tive divergence, this departure would stem from large-scale or regional advection effects,
involving horizontal variation of surface or atmospheric conditions.
Subsequent investigations have shown, however, that over wet surfaces, true equilib-
rium conditions are encountered only rarely, ifever. The main reason for this is that the
atmospheric boundary layer is never a perfectly homogeneous boundary layer, as would
be the case in channel flow; rather, it is continually responding to unsteady large-scale
weather patterns, involving condensation aloft and dry air entrainment, which tend to
maintain a humidity deficit even over the ocean. Nevertheless, the idea underlying Equa-
tion (4.30) has led Priestley and Taylor (1972) to use equilibrium evaporation as the basis
for an empirical relationship to describe evaporation from a wet surface under conditions
of minimal advection,
E
pe
. With data obtained over ocean and moist land surfaces they
concluded that it is roughly proportional to
E
e
, that is
+
γ
E
pe
=
α
e
Q
ne
(4.31)
where
α
e
is a constant, which they found to be about 1.26. This value was later confirmed
in many other studies (see Brutsaert, 1982) and
α
e
is now generally accepted to be of the
order of 1.20-1.30, on average, for advection-free water surfaces and moist landsurfaces
with short vegetation. Equation (4.31) is equivalent with a Bowen ratio
Bo
pe
=
α
−
1
[(
γ/
)
+
1]
−
1
(4.32)
e
which is illustrated inFigure 4.3 for different
α
e
values, together with some experimental
data points.
These values of
α
e
indicate that over the ocean or other moist surfaces the second term
of (4.23), that is the large-scale advection, accounts on average for about 20% to 23% of
the evaporation rate. But this is only an average and large variations have been observed in
different experimental settings. Still, it is remarkable that so many landsurfaces covered
with fairly short vegetation, such as grass, which is not actually wet but with ample water
available to the roots, yield about the same average values, ranging between 1.20 and
1.30, as open water surfaces. This may be the result of a fortuitous compensation of the
specific humidity of non-wet leaf surfaces, which is lower than saturation, by a larger
effective roughness, and thus transfer coefficient, of the vegetative surface. Still, in some
studies drastically different values of
α
e
have been reported. This has been especially
the case for very rough surfaces; for instance McNaughton and Black (1973) obtained
α
e
=
1.05 for a young,8mhigh fir forest.
