The use of Lanbsat digital classifications or enhancements to monitor change in semi-arid environments has generally met with limit~d success. This can be partially attributed to unique physical and human factors which complicate change detection in these environments. An enhanced classification approach which combines image enhancement to isolate chartge with multispectral classification to identify change dynamics has been developed. The technique has been applie1d in northwestern Nigeria to dry season Landsat MSS images acquired for dates before and after construction of the Bakolori dam and reservoir. These images span a nine-year period during which marked changes have occurred as ~ result of dam construction and stream regulation associated with the Bakolori project. The results show evidence of land degradation in flood plain areas downstream of the Bakolori dam where flood plain cultivation has been reducJd by as much as 50 percent. Comparisons with ground survey data confirm that the enhanced classification approach pr6vides more accurate information on change by minimizing errors associated with misregistration and misclassification and by allowing the suppression of environmental factors through the separation of natural and human-induced change.
[1]
Philip J. Howarth,et al.
Landsat digital enhancements for change detection in urban environments
,
1983
.
[2]
Philip J. Howarth,et al.
Procedures for change detection using Landsat digital data
,
1981
.
[3]
T. Frank,et al.
Assessing change in the surficial character of a semiarid environment with Landsat residual images
,
1984
.
[4]
J. R. Jensen.
Urban Change Detection Mapping Using Landsat Digital Data
,
1981
.
[5]
W. Todd.
Urban and regional land use change detected by using Landsat data
,
1977
.
[6]
John E. Estes,et al.
Deriving land use/land cover change statistics from Landsat - A study of prime agricultural land
,
1980
.