Geoscience Reference
In-Depth Information
(CLM) and estimated its impact on the climate of Amazonia and elsewhere. Based
on two simulations, Lee
et al.
showed that photosynthesis and evapotranspiration
increase significantly in the Amazon during the dry season if the hydraulic
redistribution process is included, resulting in less water stress and greater
evapotranspiration. Other important new findings from field observations, such as
the functioning of the plant roots in arid and semi-arid climates (Seyfried
et al.
,
2005) may also merit representation in SVATS, in conjunction with better
representation of the dynamics of surface water and groundwater interactions.
All the models in the groups of land surface sub-models discussed in the
previous three sections implicitly apply the one-dimensional Richards equation
when describing the soil moisture movement within the soil column, while the
lateral flows and/or interactions are represented through parameterizations and
flow routing. However, the routing scheme of Guo
et al.
(2004) is significantly
different in four respects:
1.
it allows grid-based runoff to exit modeled grid squares in multiple direc-
tions simultaneously instead of just one of the eight discrete directions
employed in most routing schemes, a feature which is likely to be most use-
ful for models with coarse grid resolution;
2.
it introduces a 'tortuosity coefficient' which adjusts some geomorphology-
related parameters such as channel slope and length, to reduce the impacts
of different spatial resolutions on flow routing;
3.
it uses a flow network which is reasonably realistic; and
4.
it explicitly differentiates between overland and river flow in the flow
network.
Choi
et al.
(2007) also propose applying a three-dimensional Richards equation to
more accurately represent both vertical and lateral flow interactions. There are
issues still to be resolved before three-dimensional approaches can be applied
effectively in SVATS, but this new direction of investigation merits attention.
On the basis of the above review of SVATS, it is clear that there has already been
substantial progress and that this field of interest remains active, with progress still
being made. It is anticipated that land surface sub-models will continue to evolve to
include, for example, better representation of surface-groundwater interactions,
sediment transport, biogeochemical processes, and sub-grid spatial variability
associated with the integrated atmosphere-vegetation-land-soil system, see, for
example, Fig. 24.9. However, given the problems associated with defining the
growing number of parameters that will need to be specified globally in such
models, it is not yet clear that further development and associated sub-model
complexity will necessarily have a major positive impact on the accuracy and
reliability with which predictions of weather and climate can be made.
What is arguably more certain is that further development may enhance the
capability to
interpret
predicted weather and climate in terms of their impact on
human welfare and ecological status because representation of features relevant to
such impact are included in the models themselves, with parameters that can be
