300
Climatic Geomorphology
the increase due to the gravitational component, therefore, sediment production dropped.
The relationship between rock fragment cover and sediment production has been studied
in plots by Simanton et al. (1984). For similar slopes, negative correlations were obtained
due to a number of factors. In laboratory experiments, however, Poessen and Lavee (1991)
found positive correlations when the clasts were greater than 5 cm in diameter. On the
other hand, erosion increases after dry periods and decreases during humid ones due to
the effect of vegetation that protects the surface from raindrop impact and reduces
surface flow.
One of the most important processes that affects inter-fill areas is that of rainsplash
that provides material that can then be mobilized by sheetflow. When a raindrop
reaches the soil (Figure 13.17), the impact generates a small crater by compression and
the jolted particles are scattered over the surface by shearing. The greater the kinetic
energy of the raindrop, the more important is the effect of rainsplash. If there is a thin
film of water on the surface, particle dispersion is increased, reaching a maximum when
the thickness of the water lamina is the same as the diameter of the water drop (Palmer,
1963). On horizontal surfaces the distance that the material is ejected decreases
exponentially from the point of impact (Savat and Poesen, 1981; Torri et al., 1987).
Rainsplash is most effective when there is an increase in slope (Ellison, 1944) and wind
speed. This modifies the trajectory of the majority of the mobilized particles. In
addition, vegetation cover, the size and percentage of surface particles, organic content,
soil chemistry and aggregate stability, amongst other factors, all exert an influence on
this process (Scoging, 1989). The rainsplash is an important factor that affects the
supply of particles to sheetflow and their subsequent transport in rill and inter-rill areas
(Bryan, 1987).
5. Water erosion on slopes
Erosion by surface flow may occur by sheetflow or filling. According to Leopold et al.
(1966), 98% of sediment production during a period of 10 years in a semi-arid area of
New Mexico, came from slopes without rills. In environments of greater precipitation
in Kenya, Dunne (1980) found no filling over long slopes, confirming field experiments
using rainfall simulation (Dunne and Dietrich, 1980). Similarly, Emmett (1978)
indicated that some slopes in Wyoming not affected by rilling suffer uniform
degradation through sheetflow erosion. Whereas rill erosion has not received much
attention, inter-rill erosion has been intensively studied by a number of authors who
have focused their investigations on the interaction of splash and sheetflow erosion, both
in the field and in the laboratory. These studies have taken into account slope and the
physical, chemical and mineralogical properties of soils. They indicate that the processes
and rates of inter-fill erosion are extremely complicated and variable, depending on
extrinsic factors, such as rainfall intensity, raindrop size, and the presence or absence of
wind, and intrinsic factors such as soil texture, the nature of soil aggregates, surface
roughness, susceptibility to cracking, and the presence and density of organic detritus
(Bryan, 1987).
Rill erosion is predominantly produced by the mobilization of soil particles by
concentrated flow and affects a small part of the terrestrial surface. Rills may be developed
Précédent

- 309/769

Suivant