2
1 Introduction
A variety of techniques have been used to estimate the soil erosion in previous
studies, each with intrinsic limitations and uncertainties. Statistics on the flow characteristics in the process of confluence in various gullies revealed that the combination
of sediment concentration and size distribution played a critical role in maintaining
the flow with heavy sediment load owing to the natural adjustments in the Loess Hill
Ravine Region of the Yellow River Basin (Wang et al. 1982). With a similar method,
Hovius et al. (1997) also found that sediments discharged from the western Southern
Alps were dominated by landslide-derived materials. The strain probe method was
used to continuously detect soil creep in situ, and then the volume of slope failure was
accordingly pre-estimated (Iverson et al. 2000; Yamada 1999). The tracer element
method is a helpful approach (Wen et al. 2003). Stereo photogrammetry was also
applied to determine the surface movement of the failed landslide mass (Ochiai et al.
2004). Recently, remote sensing technologies, including terrestrial laser scanning
(Oppikofer et al. 2008), sonar bathymetry (Haflidason et al. 2005), radar altimetry
(Velicogna and Wahr 2006), aerial photography (Martin et al. 2002; Whitehouse
1983), approach combing aerial photograph with satellite imagery (Li et al. 2013),
have been used to monitor soil erosion and geomorphic evolution. However, it is
very difficult to monitor the time-varying process of an individual mass failure with
these techniques because of the randomness and suddenness of such an event.
Rainfall simulation is widely used for the experimental study on soil and water
conservation. The simulators could be divided into two main groups (Cheng et al.
2008): the non-pressurized rainfall simulators, including the “thread droppers” and
the “needle droppers” with drop formers primarily made of wool fibers or hypodermic
needles, and the pressurized water rainfall simulators, including the “spout” and the
“sprayer”, of which raindrops sprayed out of a row of spouts on the pipe or a nozzle
under pressure. The former could be hardly found at present due to its drawback that
the minimum size of drops produced is far larger and more even than most of the
natural rainfalls and the simulator could not easily generate rains with kinetic energy
similar to the natural ones. Rainfall simulation halls with automatic operation and
observation systems have been booming in recent years, but the basic simulators will
not be completely replaced in virtue of the cheap price and convenient manipulation.
Nevertheless, the main limitation of the experimental method is probably linked
with the major difficulty to transfer the experimental results by analogy to larger
landforms (Schumm et al. 1987). Confirmation of the experimental results must come
from field investigations, and frequently this can be done by considering existing
field data from a new perspective that has been provided by the observation of the
small landforms as they evolve and react. Hence, what we could do is to simulate
the landslide processes by employing a conceptual slope under rainfall simulation.
However, the representative catchments on the Chinese Loess Plateau for scientific
research are generally far from urban areas, lack water and electricity, and have deep
ditches, steep slopes, inconvenient transportation, and strong winds. The special
situation poses another severe challenge to the site survey. Here a mobile shelter is
recommended in the book that could be assembled in the field to conduct site-specific
tests based on the terrain of the Loess Plateau.
1 Introduction
A variety of techniques have been used to estimate the soil erosion in previous
studies, each with intrinsic limitations and uncertainties. Statistics on the flow characteristics in the process of confluence in various gullies revealed that the combination
of sediment concentration and size distribution played a critical role in maintaining
the flow with heavy sediment load owing to the natural adjustments in the Loess Hill
Ravine Region of the Yellow River Basin (Wang et al. 1982). With a similar method,
Hovius et al. (1997) also found that sediments discharged from the western Southern
Alps were dominated by landslide-derived materials. The strain probe method was
used to continuously detect soil creep in situ, and then the volume of slope failure was
accordingly pre-estimated (Iverson et al. 2000; Yamada 1999). The tracer element
method is a helpful approach (Wen et al. 2003). Stereo photogrammetry was also
applied to determine the surface movement of the failed landslide mass (Ochiai et al.
2004). Recently, remote sensing technologies, including terrestrial laser scanning
(Oppikofer et al. 2008), sonar bathymetry (Haflidason et al. 2005), radar altimetry
(Velicogna and Wahr 2006), aerial photography (Martin et al. 2002; Whitehouse
1983), approach combing aerial photograph with satellite imagery (Li et al. 2013),
have been used to monitor soil erosion and geomorphic evolution. However, it is
very difficult to monitor the time-varying process of an individual mass failure with
these techniques because of the randomness and suddenness of such an event.
Rainfall simulation is widely used for the experimental study on soil and water
conservation. The simulators could be divided into two main groups (Cheng et al.
2008): the non-pressurized rainfall simulators, including the “thread droppers” and
the “needle droppers” with drop formers primarily made of wool fibers or hypodermic
needles, and the pressurized water rainfall simulators, including the “spout” and the
“sprayer”, of which raindrops sprayed out of a row of spouts on the pipe or a nozzle
under pressure. The former could be hardly found at present due to its drawback that
the minimum size of drops produced is far larger and more even than most of the
natural rainfalls and the simulator could not easily generate rains with kinetic energy
similar to the natural ones. Rainfall simulation halls with automatic operation and
observation systems have been booming in recent years, but the basic simulators will
not be completely replaced in virtue of the cheap price and convenient manipulation.
Nevertheless, the main limitation of the experimental method is probably linked
with the major difficulty to transfer the experimental results by analogy to larger
landforms (Schumm et al. 1987). Confirmation of the experimental results must come
from field investigations, and frequently this can be done by considering existing
field data from a new perspective that has been provided by the observation of the
small landforms as they evolve and react. Hence, what we could do is to simulate
the landslide processes by employing a conceptual slope under rainfall simulation.
However, the representative catchments on the Chinese Loess Plateau for scientific
research are generally far from urban areas, lack water and electricity, and have deep
ditches, steep slopes, inconvenient transportation, and strong winds. The special
situation poses another severe challenge to the site survey. Here a mobile shelter is
recommended in the book that could be assembled in the field to conduct site-specific
tests based on the terrain of the Loess Plateau.