8
2 Similarity of Model Experiments
the specific materials. Both of the natural or simulated rainfall, mostly the simulated
one, may be applied in a model experiment. The laboratory experiments include
full-scale models and variational-scale models. In a full-scale model experiment, the
geometry scale of the underlying surface is consistent with the prototype, while in a
variational-scale model, the landform is scaled down based on the prototype. A successful physical model experiment requires similar necessary conditions. However,
the model experiment may be conducted in the laboratory far away from the study
area, and the experimental scenario may be freely designed if needed. Hence, the
model experiment is recommended as an important supplement of the field experiment and some advantages of the experiment in the laboratory cannot be substituted
with field experiments.
These can be categorized into field experiments and model experiments according to the formation of the underlying surface of the experimental area. Based on
characteristics of the watershed geomorphology, runoff and sediment distribution
under natural and simulated rainfall conditions or other dynamic conditions of erosion, field experiments are conducted to study the mechanism of soil and water loss
and the effects of management measures on soil and water conservation, such as
the small-plot runoff experiments. It is easy to simulate and observe the complicated
topographical conditions of soil and water loss in the field, because there is no change
in the scale of the underlying surface and properties of erosion material, which means
that to reconstruct the plot is not needed. Based on geomorphology and geology of
the natural watershed, model experiments are conducted to study the mechanism of
soil and water loss and effects of various management measures for the prototype
watershed under the natural or simulated rainfall events (mostly simulated rainfall
events) by reconstructing the plot using certain materials. Model experiments include
the full-scale model experiments and the scale model experiments. For a full-scale
model experiment, the geometry scale of underlying surface is consistent with its
prototype, while the scale model experiment is designed based on the zoom scale of
prototype. A successful physical model experiment requires similar necessary conditions. However, model experiments can be conducted in the laboratory far away
from the soil loss field and assumed with an unlimited scale of the underlying surface.
Therefore, some features of the model experiments cannot be replaced by those of
the field experiments, and the model experiments are recommended as an important
supplement to the field experiments.
Under laboratory conditions, measurements are more accurate and many experiments can be conducted (Cerdà and García-Fayos 2002). Thus, downscaled models
are currently common in many different engineering fields, such as hydraulics and
river engineering. The advantages of such models are well known (Zhang 1994).
However, few studies have simulated the process of soil loss using downscaled model
experiments, because simulating the hydrological, morphological, and geological
conditions is extremely complex. Nevertheless, dimensional analysis links various
observed phenomena of erosion and deposition into a unified process, and enables
complete predictions in the landfrom changes to be anticipated when watershed treatment is altered (Strabler 1958). Downscaled models are typically used to simulate
physiognomy performance (Jin et al. 2003; Hancock and Willgoose 2003). Hancock
2 Similarity of Model Experiments
the specific materials. Both of the natural or simulated rainfall, mostly the simulated
one, may be applied in a model experiment. The laboratory experiments include
full-scale models and variational-scale models. In a full-scale model experiment, the
geometry scale of the underlying surface is consistent with the prototype, while in a
variational-scale model, the landform is scaled down based on the prototype. A successful physical model experiment requires similar necessary conditions. However,
the model experiment may be conducted in the laboratory far away from the study
area, and the experimental scenario may be freely designed if needed. Hence, the
model experiment is recommended as an important supplement of the field experiment and some advantages of the experiment in the laboratory cannot be substituted
with field experiments.
These can be categorized into field experiments and model experiments according to the formation of the underlying surface of the experimental area. Based on
characteristics of the watershed geomorphology, runoff and sediment distribution
under natural and simulated rainfall conditions or other dynamic conditions of erosion, field experiments are conducted to study the mechanism of soil and water loss
and the effects of management measures on soil and water conservation, such as
the small-plot runoff experiments. It is easy to simulate and observe the complicated
topographical conditions of soil and water loss in the field, because there is no change
in the scale of the underlying surface and properties of erosion material, which means
that to reconstruct the plot is not needed. Based on geomorphology and geology of
the natural watershed, model experiments are conducted to study the mechanism of
soil and water loss and effects of various management measures for the prototype
watershed under the natural or simulated rainfall events (mostly simulated rainfall
events) by reconstructing the plot using certain materials. Model experiments include
the full-scale model experiments and the scale model experiments. For a full-scale
model experiment, the geometry scale of underlying surface is consistent with its
prototype, while the scale model experiment is designed based on the zoom scale of
prototype. A successful physical model experiment requires similar necessary conditions. However, model experiments can be conducted in the laboratory far away
from the soil loss field and assumed with an unlimited scale of the underlying surface.
Therefore, some features of the model experiments cannot be replaced by those of
the field experiments, and the model experiments are recommended as an important
supplement to the field experiments.
Under laboratory conditions, measurements are more accurate and many experiments can be conducted (Cerdà and García-Fayos 2002). Thus, downscaled models
are currently common in many different engineering fields, such as hydraulics and
river engineering. The advantages of such models are well known (Zhang 1994).
However, few studies have simulated the process of soil loss using downscaled model
experiments, because simulating the hydrological, morphological, and geological
conditions is extremely complex. Nevertheless, dimensional analysis links various
observed phenomena of erosion and deposition into a unified process, and enables
complete predictions in the landfrom changes to be anticipated when watershed treatment is altered (Strabler 1958). Downscaled models are typically used to simulate
physiognomy performance (Jin et al. 2003; Hancock and Willgoose 2003). Hancock