Mapping of Colophospermurn mopane using Landsat TM in eastern Botswana

The spatial distribution of a tree species, Colophospermum mopane is mapped at its southern limits in eastern Botswana. Its relationship with soil type distribution is also investigated. Methods applied in the exercise used the possibilities of integrated applications of geographic information syste...

Full description

Saved in:
Bibliographic Details
Published in:South African geographical journal 2008, Vol.90 (1), p.41
Main Authors: Sebego, Reuben J., Arnberg, Wolter, Lunden, Bengt, Ringrose, Susan
Format: Article
Language:English
Subjects:
cover changes
Earth sciences
Geovetenskap
imagery
management
NATURAL SCIENCES
NATURVETENSKAP
reflectance
remotely-sensed data
vegetation
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The spatial distribution of a tree species, Colophospermum mopane is mapped at its southern limits in eastern Botswana. Its relationship with soil type distribution is also investigated. Methods applied in the exercise used the possibilities of integrated applications of geographic information systems and remote sensing to detect and delineate spectral classes of C. mopane and of soils. Landsat TM data for the wet and dry seasons were analysed and interpreted for soil and vegetation classification. Supervised classification of the wet season (19/04/93) image was carried out using land use/cover classes derived from Colour infrared (CIR) aerial photographs, flown during the wet season (April, 1996) over the study area. Classification of the dry season (19/10/84) image on the other hand was based on the existing Food and Agricultural Organisation (FAO) soil maps of the same area. Results show that CIR photographs can be reliably used in selecting training areas for the classification of Landsat TM imagery. Secondly, results also indicate that Landsat TM can be used to map the distribution pattern of mopane. The methods applied in this study and the tools used prove that the much needed inventorying of vegetation in Botswana can be carried out using remote sensing. Lastly, the study shows that spectral classes of soil and vegetation have a positive relationship.
ISSN:0373-6245
2151-2418
DOI:10.1080/03736245.2008.9725310