both run-off and soil loss are monitored. Plot sizes are in the order of 20 m length
and 2–4 m width, although other plot sizes are sometimes used (Morgan 2005). Plot
edges are usually made of sheet metal, wood or any material (i.e., concrete) that is
stable and does not leak, and is not liable to rust. The edges should extend
0.15–0.20 m above the soil surface and be embedded in the soil to prevent run-on
water intrusion. At the downslope end of an individual plot is a collecting trough or
gutter from which sediment and run-off are channeled into collecting tanks.
Bounded run-off plots give probably the most reliable data on soil loss per unit
area. However, up-scaling the results for watershed level assessments can lead to
severe over- or under-estimation due to the uncontrolled field situation, including
having to disregard boundary and slope length effects as well as variability/unevenness of soil surface conditions at the larger scale (Bonell and Bruijnzeel 2005).
In contrast, the aim of unbounded plots is to determine soil erosion or sediment
transport rates for larger areas such as hills or watersheds under local management
conditions (Morgan 2005). This type of monitoring plot offers the opportunity to
assess the local soil loss and sediment transport magnitude over a greater spatial
scale, which can be seen as a viable alternative both to the bounded plot approach
and controlled field conditions. In the case of unbounded plots, soil erosion can be
measured by i.e., sediment fences (Robinchaud and Brown 2002), which usually
consist of a filter fabric stretched across a slope or foothill area and attached to
supporting posts. The fence should be located in areas where run-off waterways
pass by, in order to capture transported sediments for soil texture analysis and to
determine the amount of transported material. Different types of fiber are
commonly used, such as polyester plastics or linen, but all types must be able to
withstand water pressure during run-off events and retain sandy to clay soil
particles, implying a low permeability of these particles by the fabric. Sediment
fences are not practical where large flows of water are involved, and have to be
renewed after every wet season in the case of tropical environments, or in even
shorter timespans depending on local rainfall severity, such as after tropical storms
and erratic rainfall events. In general, the use of sediment fences is recommended
for small drainage areas only (<1 ha), and problems can arise due to the incorrect
selection of filter fabrics or improper installation (USEPA 1992) or the inexact
delineation of water flow pathways. In contrast to this medium-scale monitoring
technique, turbidity sensors (Levis 1996) are a viable monitoring option for the
determination of the severity of soil erosion in larger watersheds or basins. These
sensors are used to continuously monitor suspended sediment concentrations (SSC)
in rivers or streams; for example, at a watershed outlet. Given appropriate calibration, they even have the potential to determine carbon and nitrogen fluxes. Turbidity is generally a much better predictor than water discharge in terms of SSC,
turbidity being the optical measure of the cloudiness of water, as caused by light
being scattered by suspended particles, organic matter and dissolved constituents
(Levis 1996).
7 Soil Conservation on Sloping Land: Technical Options and Adoption Constraints
253
and 2–4 m width, although other plot sizes are sometimes used (Morgan 2005). Plot
edges are usually made of sheet metal, wood or any material (i.e., concrete) that is
stable and does not leak, and is not liable to rust. The edges should extend
0.15–0.20 m above the soil surface and be embedded in the soil to prevent run-on
water intrusion. At the downslope end of an individual plot is a collecting trough or
gutter from which sediment and run-off are channeled into collecting tanks.
Bounded run-off plots give probably the most reliable data on soil loss per unit
area. However, up-scaling the results for watershed level assessments can lead to
severe over- or under-estimation due to the uncontrolled field situation, including
having to disregard boundary and slope length effects as well as variability/unevenness of soil surface conditions at the larger scale (Bonell and Bruijnzeel 2005).
In contrast, the aim of unbounded plots is to determine soil erosion or sediment
transport rates for larger areas such as hills or watersheds under local management
conditions (Morgan 2005). This type of monitoring plot offers the opportunity to
assess the local soil loss and sediment transport magnitude over a greater spatial
scale, which can be seen as a viable alternative both to the bounded plot approach
and controlled field conditions. In the case of unbounded plots, soil erosion can be
measured by i.e., sediment fences (Robinchaud and Brown 2002), which usually
consist of a filter fabric stretched across a slope or foothill area and attached to
supporting posts. The fence should be located in areas where run-off waterways
pass by, in order to capture transported sediments for soil texture analysis and to
determine the amount of transported material. Different types of fiber are
commonly used, such as polyester plastics or linen, but all types must be able to
withstand water pressure during run-off events and retain sandy to clay soil
particles, implying a low permeability of these particles by the fabric. Sediment
fences are not practical where large flows of water are involved, and have to be
renewed after every wet season in the case of tropical environments, or in even
shorter timespans depending on local rainfall severity, such as after tropical storms
and erratic rainfall events. In general, the use of sediment fences is recommended
for small drainage areas only (<1 ha), and problems can arise due to the incorrect
selection of filter fabrics or improper installation (USEPA 1992) or the inexact
delineation of water flow pathways. In contrast to this medium-scale monitoring
technique, turbidity sensors (Levis 1996) are a viable monitoring option for the
determination of the severity of soil erosion in larger watersheds or basins. These
sensors are used to continuously monitor suspended sediment concentrations (SSC)
in rivers or streams; for example, at a watershed outlet. Given appropriate calibration, they even have the potential to determine carbon and nitrogen fluxes. Turbidity is generally a much better predictor than water discharge in terms of SSC,
turbidity being the optical measure of the cloudiness of water, as caused by light
being scattered by suspended particles, organic matter and dissolved constituents
(Levis 1996).
7 Soil Conservation on Sloping Land: Technical Options and Adoption Constraints
253