Remote sensing based multi-temporal land cover classification and change detection in northwestern ethiopia

Research output: Contribution to journal › Research article › Contributed › peer-review

Contributors

  • Worku Zewdie - , TUD Dresden University of Technology (Author)
  • Elmar Csaplovics - , Chair of Remote Sensing (Author)

Abstract

Spatiotemporal change analysis of semiarid regions is vital for understanding major threats to the ecosystem. This study examines land use and land cover (LULC) changes using multitemporal satellite imagery for the period 1972-2010. Supervised classification algorithm using support vector machines (SVM) was employed to monitor LULC transformations. A cross-tabulation matrix was used to assess the total change of land categories based on net change and swap change. The major land use change in this dynamic region were conversion of about 52% woodlands to intensive land uses such as cropland in the period 1972 - 2010. The net change of woodland accounts for over 61% and net-gain of cropland and grassland were about 53% and 9% respectively. Based on the socio-ecological field survey expansion of croplands, population pressure as well as overharvesting of trees, respectively, are major drivers of change. This significant change in land use is mainly due to accelerated human impact and subsequent agricultural land expansion.The result of this study provides a vital monitoring basis for continuous investigations of changes in the natural vegetation of semiarid environments.

Details

Original languageEnglish
Article numberA008
Pages (from-to)121-139
Number of pages19
JournalEuropean Journal of Remote Sensing
Volume48
Issue number1
Publication statusPublished - 2015
Peer-reviewedYes

Keywords

Sustainable Development Goals

Keywords

  • Landsat Imagery, Net change, Persistence, Semi-arid, Support vector machines (SVM), Swap change