7 Land-use and Catchment Characteristics
135
tion between optical imagery in (a) and (b), vs. radar imagery in (c) is implicit. It
was felt that integration of both kinds of imagery would be artificial in the context
of vegetation indices and classification. Hydrologic model parameter assessment
is particularly important for water management in changing environments (Chap.
19). Therefore, accurate assessment of parameters, as addressed in the current chapter, gains importance in multi-temporal cases. When comparing results of different
dates, observed 'changes' due to errors and incompatibilities should not be mistaken
as real changes in the parameters.
Case study area
Various techniques described in this chapter are illustrated using multi-temporal
Landsat TM imagery of the Rio Verde do Mato Grasso area, located at the Eastern
boundary of the Pantanal region, Mato Grosso do SuI, Brazil.
During the last two decades important transformations in land cover took place
mainly in the Planalto, where the native vegetation (shrubs and forest) was mostly
replaced by intensive cultivation methods. The systematic deforestation involved denudation of soils and caused rapid erosion, with consequences for the flood regime
in the whole Pantanal, where also an increased sedimentation rate was detected
(Hernandez Filho et aI., 1995). The study area represents a good example where
environmental dynamics linked to human activities have a strong impact on water
management.
7.3 Vegetation Indices
The relevance of vegetation for hydrology was addressed in the introduction section.
Quantified biophysical parameters from remote sensing are associated with irrigation management (Bastiaansen, 1998) (Table 7.1). The relation between remotely
sensed measurements and vegetation parameters is captured in vegetation indices.
The interest in assessing vegetation growth and conditions from space dates back
to 1972, when food crops were considered a strategic commodity, but the U.S. government was unaware of the disastrous crop situation in the Soviet Union (Calder,
1991). Crop yield prediction using satellites obtained political priority and funding
for earth observation satellite programs was secured.
Field measurement of many crops revealed that a very specific reflectance characteristic occurs in the red/near-infrared part of the electro-magnetic spectrum. Visible
light is mostly absorbed by vegetation. Even green reflectance is rather low, compared to most other materials (soils, rocks) that cover the earth's surface. At the
same time, the near-infrared reflection of healthy vegetation is much higher than
that of most other land covers. Reflectance curves of vegetation (for example Fig.
7.1) show a very steep ascent between visible red and near-infrared wavelengths
(Tucker, 1979).
135
tion between optical imagery in (a) and (b), vs. radar imagery in (c) is implicit. It
was felt that integration of both kinds of imagery would be artificial in the context
of vegetation indices and classification. Hydrologic model parameter assessment
is particularly important for water management in changing environments (Chap.
19). Therefore, accurate assessment of parameters, as addressed in the current chapter, gains importance in multi-temporal cases. When comparing results of different
dates, observed 'changes' due to errors and incompatibilities should not be mistaken
as real changes in the parameters.
Case study area
Various techniques described in this chapter are illustrated using multi-temporal
Landsat TM imagery of the Rio Verde do Mato Grasso area, located at the Eastern
boundary of the Pantanal region, Mato Grosso do SuI, Brazil.
During the last two decades important transformations in land cover took place
mainly in the Planalto, where the native vegetation (shrubs and forest) was mostly
replaced by intensive cultivation methods. The systematic deforestation involved denudation of soils and caused rapid erosion, with consequences for the flood regime
in the whole Pantanal, where also an increased sedimentation rate was detected
(Hernandez Filho et aI., 1995). The study area represents a good example where
environmental dynamics linked to human activities have a strong impact on water
management.
7.3 Vegetation Indices
The relevance of vegetation for hydrology was addressed in the introduction section.
Quantified biophysical parameters from remote sensing are associated with irrigation management (Bastiaansen, 1998) (Table 7.1). The relation between remotely
sensed measurements and vegetation parameters is captured in vegetation indices.
The interest in assessing vegetation growth and conditions from space dates back
to 1972, when food crops were considered a strategic commodity, but the U.S. government was unaware of the disastrous crop situation in the Soviet Union (Calder,
1991). Crop yield prediction using satellites obtained political priority and funding
for earth observation satellite programs was secured.
Field measurement of many crops revealed that a very specific reflectance characteristic occurs in the red/near-infrared part of the electro-magnetic spectrum. Visible
light is mostly absorbed by vegetation. Even green reflectance is rather low, compared to most other materials (soils, rocks) that cover the earth's surface. At the
same time, the near-infrared reflection of healthy vegetation is much higher than
that of most other land covers. Reflectance curves of vegetation (for example Fig.
7.1) show a very steep ascent between visible red and near-infrared wavelengths
(Tucker, 1979).