Soil erosion on interrill areas suc

Soil erosion on interrill areas such as row sideslopes is a major
source of sediment for land without soil cover (Meyer and Harmon,
1989) and has been identified as one of the major processes
contributing to soil and water quality degradation. Interrill erosion
can be described as a combination of two sub-processes: splash and
wash dynamics. These two distinct sub-processes were considered in
some erosion models such as LISEM (Limburg Soil Erosion Model) (De
Roo, 1996), EUROSEM (EUROpean Soil Erosion Model) (Morgan et al.,
1998) and TEST (Terrace Erosion and Sediment Transport model)
(Van Dijk and Bruijnzeel, 2003) and have been investigated in some
studies. Luk (1979) measured total splashed loss and wash loss from
plots of 30.5×30.5 cm2 and found that total splash loss was greater
than wash loss. Mermut et al. (1997) also found that total splashed
loss was much higher than wash loss when a cylindrical soil tray of
30 cm diameter was used. The studies of Sutherland et al. (1996) and
Wan et al. (1996) found that both downslope splash and total splash
(lateral+down+top) were greater than wash on the slopes from 5°
to 20°. Van Dijk et al. (2003a) conducted the experiments on the
slopes of 0°, 5°, 15° and 40° under natural rainfall and also obtained
that both downslope splash and total splash (lateral+down+top)
were higher than wash. However, some published data indicate that
splash loss was less than wash loss in interrill areas. Bryan and Luk(1981) partitioned downslope splash from wash and found that
downslope splash was less than wash on 12.5° slope. Fox and Bryan
(1999) conducted the simulated rainfall experiments in a
100×40×10 cm3 soil tray at the five slopes (2.5%, 11.5%, 20.5%, 30%,
and 40%) and only downslope splashed material was measured. Their
results showed that downslope splash erosion never accounted for
more than 20% of the total erosion. The studies by Bryan (1974)
partitioned total splash (lateral+front+top) from wash on a slope of
20° and found that total splash was less than wash. Bryan (1979)
measured the downslope splashed material of eight soils at ten slopes
from 3° to 30° and proved that the magnitude of splash and wash
contribution to soil erosion varied between different soil types. The
splash loss was higher than wash for grey luvisol, but less than wash
for calcareous loess. The above results indicate that relative
importance of splash and wash loss varied with tray scale, soil
properties and rainfall characteristic, etc. Generally splash loss
decreased with increase in plot size and was close to zero when
plot length was greater than 10 m (Van Dijk, 2005).
Slope is a major effect factor in soil particle detachment and
transport, particularly in interrill areas. Three alternative types of
relationships between wash loss and the slope gradient have been
demonstrated: (1) Wash loss has been predicted to be a power
function of percent of the slope gradient (Neal, 1938; Zingg, 1940;
Musgrave, 1947; Zhang et al., 1998). Exponents for data grouping
both interrill and rill erosion generally ranged from approximately 0.7
to 2 (Bryan, 1979). (2) Wash loss has been predicted to be a linear
function of the sine of the slope gradient (McCool et al., 1987; Liu et
al., 1994). The coefficient varied with the range of slope gradient. (3)