74
N.M. Mattikalli and E.T. Engman
A DEM can also be employed in conjunction with satellite reflectance data to
estimate and model hydrological processes. Dubayah (1992) employed aDEM,
Landsat TM data and a radiative transfer algorithm to model spatial variability of
net solar radiation at a fine spatial resolution. Most of the examples presented in
this chapter integrate DEM derived data with remotely sensed information for
various applications.
Remotely sensed data (viz. land-use information), DEM (slope data) and digitized soil map can be merged to determine Hydrological Similar Units (HSU),
which sub-divide a drainage basin into areas that have hydrologically similar behavior (Schultz, 1994). Su et aI. (1992) utilized overlay operation on input layers
to derive HSU for the Nims River basin in Germany (Color Plate 4.A).
Realistic 3D perspective views can be generated using a DEM to visualize terrain variation. Colour Plate 4.B shows a 3D view of elevation variation across the
Little Washita watershed, Oklahoma. Remotely sensed images can be integrated
with such 3D views for improved understanding of spatial and temporal variability
of certain hydrological parameters. Colour Plate 4.C illustrates daily sequence of
spatial distribution of remotely sensed microwave brightness temperature for the
Little Washita watershed during June 10-18, 1992. These data are useful to derive
near-surface soil moisture and relate the spatial and temporal variations to soil
properties (Mattikalli et aI., 1998).
4.3.3
Land-use! Land-cover Change Detection
Remote sensing offers multi-temporal repetitive data for identification and quantification of land surface changes, and therefore, greatly enhances capability of a
GIS in updating map information on a regular basis (Eckhardt et aI., 1990).
Michalak (1993) reviews current examples of integrated approach to land-use
change analysis. Mattikalli (1995) reports a Boolean-logic technique applied to
vector formatted layers for automatic analysis of historical land-use dynamics.
This technique performs overlay operation to merge land-use data acquired on two
dates, and then carries out Boolean operations to generate change map and associated statistics. Such a methodology is useful to analyze large amounts of data
derived from remote sensing in conjunction with information derived from maps
and aerial photographs archived in a GIS.
An expert system's approach to change detection has been implemented (Wang
and Newkirk, 1987). Automation of integrated GIS using an expert system involves three separate tasks, viz. classification of remotely sensed imagery, detection of change, and expert rules for updating the GIS database. Currently, no
commercially available GIS offers such a level of integration. However, there are
several experimental applications incorporating some of the principles of automated land-use change analysis.
Examples of land-use change detection with the aid of multi-temporal Landsat
imagery are presented in Chap. 19.
N.M. Mattikalli and E.T. Engman
A DEM can also be employed in conjunction with satellite reflectance data to
estimate and model hydrological processes. Dubayah (1992) employed aDEM,
Landsat TM data and a radiative transfer algorithm to model spatial variability of
net solar radiation at a fine spatial resolution. Most of the examples presented in
this chapter integrate DEM derived data with remotely sensed information for
various applications.
Remotely sensed data (viz. land-use information), DEM (slope data) and digitized soil map can be merged to determine Hydrological Similar Units (HSU),
which sub-divide a drainage basin into areas that have hydrologically similar behavior (Schultz, 1994). Su et aI. (1992) utilized overlay operation on input layers
to derive HSU for the Nims River basin in Germany (Color Plate 4.A).
Realistic 3D perspective views can be generated using a DEM to visualize terrain variation. Colour Plate 4.B shows a 3D view of elevation variation across the
Little Washita watershed, Oklahoma. Remotely sensed images can be integrated
with such 3D views for improved understanding of spatial and temporal variability
of certain hydrological parameters. Colour Plate 4.C illustrates daily sequence of
spatial distribution of remotely sensed microwave brightness temperature for the
Little Washita watershed during June 10-18, 1992. These data are useful to derive
near-surface soil moisture and relate the spatial and temporal variations to soil
properties (Mattikalli et aI., 1998).
4.3.3
Land-use! Land-cover Change Detection
Remote sensing offers multi-temporal repetitive data for identification and quantification of land surface changes, and therefore, greatly enhances capability of a
GIS in updating map information on a regular basis (Eckhardt et aI., 1990).
Michalak (1993) reviews current examples of integrated approach to land-use
change analysis. Mattikalli (1995) reports a Boolean-logic technique applied to
vector formatted layers for automatic analysis of historical land-use dynamics.
This technique performs overlay operation to merge land-use data acquired on two
dates, and then carries out Boolean operations to generate change map and associated statistics. Such a methodology is useful to analyze large amounts of data
derived from remote sensing in conjunction with information derived from maps
and aerial photographs archived in a GIS.
An expert system's approach to change detection has been implemented (Wang
and Newkirk, 1987). Automation of integrated GIS using an expert system involves three separate tasks, viz. classification of remotely sensed imagery, detection of change, and expert rules for updating the GIS database. Currently, no
commercially available GIS offers such a level of integration. However, there are
several experimental applications incorporating some of the principles of automated land-use change analysis.
Examples of land-use change detection with the aid of multi-temporal Landsat
imagery are presented in Chap. 19.