Environmental Engineering Reference
In-Depth Information
in place of evapotranspiration; one of the equa-
tions described in
Section 2.4.2
is then used to
estimate potential evapotranspiration. Runoff is
often estimated with the US Soil Conservation
Service curve number method (see
Section 2.4.4
),
and change in storage is determined by a book-
keeping procedure based on field capacity, wilt-
ing point capacity, and initial amount of water
in storage. Additional data may be required in
models that use more detailed water-budget
equations. Soil-vegetation-atmosphere trans-
fer (SVAT) models, for example, may expli-
citly account for plant growth as well as water
uptake from soil by plant roots and release of
that water to the atmosphere through leaf sto-
mata (e.g. Fischer
et al
.,
2008
). This more sophis-
ticated approach requires data such as plant
type, stage of growth, slope and aspect of land
surface, relative humidity, solar radiation, and
wind speed. Storage can be simulated more
accurately if information on water retention
properties of the soil is available. Similarly,
precipitation intensity and surface roughness
information can improve runoff predictions.
Land-surface models (
Section 2.5
) are forms of
soil water-budget models that act as subcompo-
nents in large-scale atmospheric general circu-
lation models.
Most models consider the unsaturated zone
to be a transition zone within which water
moves downward to the water table. Some
modeling approaches do not consider water
movement through the unsaturated zone; they
simply refer to drainage from the soil zone
as potential recharge or net infiltration (Flint
and Flint,
2007
). Others assume that drainage
from the soil zone arrives instantaneously at
the water table (Batelaan and De Smedt,
2007
;
Dripps and Bradbury,
2007
), an assumption
appropriate for sites with shallow water tables
or for studies that span long periods, such as
decades or centuries, where travel time to the
water table is negligible relative to the length of
the study period.
Water movement through the unsaturated
zone has been simulated with approaches that
range from bucket models to numerical mod-
els that solve the Richards equation. The more
detailed models offer insights into physical
processes that affect water movement, but they
require information on model parameters and
can be computationally intensive. In contrast,
bucket models require little information on the
unsaturated zone, and their application is rela-
tively easy and straightforward. These simpler
models are amenable to use with geographical
information system (GIS) applications, in which
many one-dimensional simulations are used
to represent the response of a watershed to
changes in factors such as land use, climate, or
water use.
Simple bucket models represent the sur-
face and subsurface by a series of storage res-
ervoirs that have been referred to as buckets,
tanks, compartments, or layers (
Figure 3.2
).
The shallowest subsurface bucket is filled with
infiltration from precipitation or irrigation. As
one bucket is filled, excess water flows to the
next deepest bucket, and so on. Water can be
extracted from shallow buckets by evaporation,
plant transpiration, interflow, and other phe-
nomena. Flow of water to or from the deepest
bucket is assumed to be recharge. Ragab
et al
.
(
1997
) and Finch (
1998
) proposed models that
had four layers of buckets within the root zone.
Evaporation
and
Tr anspiration
Precipitation
and
Irrigation
Runoff
Surface
Layer
Tr anspiration
Subsurface
Layer
Interflow
Recharge
Groundwater
Layer
Base Flow
Figure 3.2
Schematic diagram of simple tank model.
