Geoscience Reference
In-Depth Information
6.2
Evapo-Transpiration and Net Precipitation
6.2.1
Evapo-transpiration
Evapo-transpiration (ET; the combination of evaporation from the surface
and transpiration from plants) is equivalent to the latent heat flux examined in
Chapter 5
, but in different units; namely, water depth over a selected time period.
Assuming liquid to vapor energy transformations (or vice versa), the relationship
is as follows:
ET = (L/ ρ L
v
) t
(6.1)
where L is the latent heat flux, L
v
is the latent heat of evaporation (see
Chapter 5
)
and ρ is the density of water (which also varies slightly with respect to temper-
ature) and t is time. If the desired value of ET is mm per day, and one assumes
a water density of 10
3
kg m
−3
and the latent heat of evaporation at 0°C, one can
simply multiply the average latent heat flux by a conversion factor of 0.03456. For
example, we saw that in July at the SHEBA site (
Figure 5.11
), there was an average
latent heat flux of about 7 W m
−2
(directed upward). From Equation
6.1
, this yields
a daily evaporation rate of 0.24 mm per day, or 7.5 mm per month. This is a rather
small value. Taking from
Table 5.4
a mid-range value for the latent heat of 50 W
m
−2
for summer months over the Alaskan tundra (based on field measurements),
yields a monthly ET nearly an order of magnitude higher. In a cold environment
where surface is completely frozen, one would replace L
v
in Equation
6.1
with the
latent heat of sublimation.
Direct estimates of the latent heat flux (and hence of ET) in the Arctic are
particularly sparse. Along with direct measurements (generally based on the pro-
file and eddy correlation methods reviewed in
Chapter 5
), estimates have been
compiled from more simple energy budget considerations (e.g., Zubenok
1976
;
Lydolph
1977
; Fisheries and Environment Canada
1978
), as a residual (e.g., Kane
et al.
1992
; Serreze et al.,
2003a
) and from Land Surface Models (LSMs) (Slater
et al.,
2001
). Residual estimates of ET can be made at the watershed scale via
the water balance equation at the surface: ET = P - R (where for a long-term
annual average the storage terms in snow cover and the soil can be neglected) or
via the atmospheric moisture balance (the “aerological method”), discussed later.
Briefly, and as described more completely in
Chapter 9
, LSMs address interac-
tions between the land surface, atmosphere, and underlying surface. Time series
of basic variables (generally downward shortwave and longwave radiation, pre-
cipitation, near-surface winds, humidity, and air temperature) represent model
forcings. The model ingests these forcings and generates numerous outputs, one
of which is ET. These forcing can variously be direct observations or outputs
from an atmospheric model (often atmospheric reanalyses) to which the LSM is
coupled. The estimated ET is only as good as the physics of the LSM and of the
driving fields.
