Paper
10 October 2014 Monitoring land use/land cover dynamics in northwestern Ethiopia using support vector machine
Worku Zewdie, E. Csaplovics
Author Affiliations +
Abstract
Land use/land cover (LULC) change assessment explores a terrestrial ecosystem in relation to the impact of natural processes and anthropogenic activities towards temporal and spatial change. This study explores spatial and quantitative dynamics of land use change in the semi-arid regions of northwestern Ethiopia using Landsat-5 (1984) and Landsat-8 (2014) which provided recent and historical LULC conditions of the region. Supervised classification algorithm using support vector machines (SVM) was used to map and monitor land use transformations. A post-classification change detection assessment was applied to individual image classification outputs of the best performing SVM model in order to identify respective two-date change trajectories. The change detection analysis with an extended transition matrix showed a net quantity change of 44.0% and total change of 53.7% of the study area, with the latter change is due to swap changes. Post-classification comparisons of the classified imagery identified a major woodland transformation to cropland which is attributed to population size and economic activity. The area of cropland has increased significantly (52.8%) in 2014 contributing to the reduction in native vegetation cover. In the study period, 55.6% of woodland lost signifying a significant change in ecosystems. This significant land use transformation is due to accelerated human impact and subsequent agricultural land expansion. The loss in vegetation cover has exposed the surface and it is common to see a haze of cloud in a most semiarid region of NW Ethiopia.
© (2014) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Worku Zewdie and E. Csaplovics "Monitoring land use/land cover dynamics in northwestern Ethiopia using support vector machine", Proc. SPIE 9245, Earth Resources and Environmental Remote Sensing/GIS Applications V, 92450W (10 October 2014); https://doi.org/10.1117/12.2066461
Lens.org Logo
CITATIONS
Cited by 2 scholarly publications.
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Earth observing sensors

Landsat

Ecosystems

Vegetation

Agriculture

Image classification

Satellites

Back to Top