12 Soil Erosion
281
(subsoil soil at the surface). They found this technique to be useful for bare soils but
severely limited under condition with significant vegetation.
Price (1993) in a study of Pinyon-Juniper woodland found that spectral measurements from TM were better indicators of soil erosion patterns than estimates from
USLE. He concluded that spectral properties of remotely sensed data gave an
integrated view of soil erosion of the landscape and that remotely sensed data should
be considered when developing soil erosion models for broad geographic areas of
semiarid and arid land.
While such simple relationship between radiance measured in a satellite band have
been related to soil erosion, most studies have used a vegetation index (i.e., NDVI,
BI) or other band ratios or combinations to estimate soil erosion areas from satellite
data. Mathieu et al. (1997) using SPOT data found that bands and band ratios
provided a reliable means of determining a soil surface affected by erosion. However,
they were not able to extrapolate these fmdings into predictive indicators of erosion.
In a study in Norway, Leek and Solberg (1995) used CIR, SPOT, and ERS-l SAR
data to monitor autumn tillage and concluded that remotely sensed data was useful for
monitoring soil exposed to rainfall soil erosion. They also found that a combination
of optical and microwave data was effective in monitoring tillage areas because the
microwave allowed data collection during cloud cover. By mapping these bare soil,
they were able to delineate areas of potential soil erosion problems.
Since surface soil erosion is highly correlated the amount of surface residue, techniques to estimate residue cover are needed to better determine soil loss. Daughtry et
al (1996) have shown that reflectance and fluorescence can be used to determine the
amount of residue on the soil surface. They concluded the fluorescence techniques
were better and have developed methods to estimate surface residue in the field.
Whiting et al. (1987) used small format photographs to estimate residue cover and
concluded that the technique was useful for monitoring conservation tillage and soil
exposed to rainfall soil erosion. They found the method could reduce field time
needed to monitor surface residue and soil erosion. Biard and Baret (1997) developed
an algorithm (CRIM - Crop Residue Index Multiband) to estimate surface residue.
They showed "reasonably good" estimates of residue using field measurements of
reflectance in the TM bands. The algorithm could be used with any set of wave bands.
Biard and Baret (1997) also provide a good background on studies on residue
measurements using spectral data from ground and satellite sensors.
12.4.4 Topographic Measurements
Topographic maps have long been used in studies of soil erosion. However, the need
for more up-to-date, accurate and rapid measurements and assessments ofland surface
terrain features to estimate land surface roughness, water movement, and soil erosion
has led to the application oflaser distancing technology for in situ measurements of
soil surfaces from a laser mounted 10-20 cm above the surface (Bertuzzi et al. 1990;
Huang and Bradford 1990). These in situ studies have been used to quantify soil loss
from rill and sheet soil erosion with high accuracy (millimeter accuracy) and to study
281
(subsoil soil at the surface). They found this technique to be useful for bare soils but
severely limited under condition with significant vegetation.
Price (1993) in a study of Pinyon-Juniper woodland found that spectral measurements from TM were better indicators of soil erosion patterns than estimates from
USLE. He concluded that spectral properties of remotely sensed data gave an
integrated view of soil erosion of the landscape and that remotely sensed data should
be considered when developing soil erosion models for broad geographic areas of
semiarid and arid land.
While such simple relationship between radiance measured in a satellite band have
been related to soil erosion, most studies have used a vegetation index (i.e., NDVI,
BI) or other band ratios or combinations to estimate soil erosion areas from satellite
data. Mathieu et al. (1997) using SPOT data found that bands and band ratios
provided a reliable means of determining a soil surface affected by erosion. However,
they were not able to extrapolate these fmdings into predictive indicators of erosion.
In a study in Norway, Leek and Solberg (1995) used CIR, SPOT, and ERS-l SAR
data to monitor autumn tillage and concluded that remotely sensed data was useful for
monitoring soil exposed to rainfall soil erosion. They also found that a combination
of optical and microwave data was effective in monitoring tillage areas because the
microwave allowed data collection during cloud cover. By mapping these bare soil,
they were able to delineate areas of potential soil erosion problems.
Since surface soil erosion is highly correlated the amount of surface residue, techniques to estimate residue cover are needed to better determine soil loss. Daughtry et
al (1996) have shown that reflectance and fluorescence can be used to determine the
amount of residue on the soil surface. They concluded the fluorescence techniques
were better and have developed methods to estimate surface residue in the field.
Whiting et al. (1987) used small format photographs to estimate residue cover and
concluded that the technique was useful for monitoring conservation tillage and soil
exposed to rainfall soil erosion. They found the method could reduce field time
needed to monitor surface residue and soil erosion. Biard and Baret (1997) developed
an algorithm (CRIM - Crop Residue Index Multiband) to estimate surface residue.
They showed "reasonably good" estimates of residue using field measurements of
reflectance in the TM bands. The algorithm could be used with any set of wave bands.
Biard and Baret (1997) also provide a good background on studies on residue
measurements using spectral data from ground and satellite sensors.
12.4.4 Topographic Measurements
Topographic maps have long been used in studies of soil erosion. However, the need
for more up-to-date, accurate and rapid measurements and assessments ofland surface
terrain features to estimate land surface roughness, water movement, and soil erosion
has led to the application oflaser distancing technology for in situ measurements of
soil surfaces from a laser mounted 10-20 cm above the surface (Bertuzzi et al. 1990;
Huang and Bradford 1990). These in situ studies have been used to quantify soil loss
from rill and sheet soil erosion with high accuracy (millimeter accuracy) and to study