Tropentag, October 5 - 7, 2011 in Bonn "Development on the margin"

Applying Change Vector Analysis Method to Detect Vegetation Regeneration and Deforestation in Southern Darfur, Sudan

Technische Universität Dresden, Institute of Photogrammetry and Remote Sensing, Germany

Abstract

The continuous extraction of wood and the conversion of forest to agriculture due to a rapid population growth (ca. 4.15 % per year) in southern Darfur State since the drought of 1985 are rapidly changing the natural vegetation cover of the Edd Al Fursan region. The main aim of the presented study was to assess the dynamics of the change of the natural vegetation cover during the period 1972 and 2008 in this area.
Multi-temporal Landsat (MSS, ETM) and ASTER data have been utilised to detect historical vegetation change using Change Vector Analysis (CVA) of Kauth- Thomas features representing brightness and greenness. The technique produced images that show the direction and the magnitude of the vegetation cover change between two dates based on the inputs of brightness and greenness. The direction of change indicate whether a landscape experienced deforestation, reforestation, or remained persistent. The magnitude indicate to what degree the change occurred.
Supervised classification has been used to quantify different land cover classes. The results of the study show that intensive deforestation as well as bare soil expansion occurred during the periods 1972-1984, 1984-1989, 1999-2008 and 1972-2008. Since the early 1970s drought conditions were present and human migration from northern to southern Darfur took place. Moreover, in recent years conflict and war pushed a very large number of the internally displaced persons (IDPs) into that area. As a result, many areas have been cleared from its forest cover either for agricultural purpose or for wood selling. The harvested wood was sold at local markets as fire wood or building material. In contrary, significant vegetation regeneration occurred during the period 1989-1999, which indicates that the environment recovered from the drought due to sufficient annual rainfall as well as due to a livelihood diversification of the local population. The study demonstrated the capability of the CVA to classify and quantify vegetation cover change in southern Darfur.

Keywords: Change detection, change vector analysis, land cover change dynamics

Contact Address: Masarra Bashir, Technische Universität Dresden, Institute of Photogrammetry and Remote Sensing, Gerokstr. 38, 01307 Dresden, Germany, e-mail: masarrabashirhotmail.com