Detecting Land Use Land Cover Change Impacted by Civil Crisis in Ivory Coast using Remote Sensing and GIS

Abstract

Civil unrest has been identified as a powerful form of collective human dynamics, which has led to major aggressive changes in land use and land cover of societies in modern history. The District of Abidjan in Ivory Coast was faced with drastic landscape change following the civil war that started early 2000 until late 2012 that led to spontaneous inflow of people from the affected areas into the capital city for protection and survival. This paper examines the impact of over a decade civil unrest on the land use and land cover changes and its implications on infrastructure and utility service delivery in the capital city of Ivory Coast using Landsat ETM+ and Landsat 8 imageries acquired in December 2002 and April 2014. First, the two images were classified using the supervised classification scheme and the accuracy of classification assessed to examine their suitability for change detection. Thereafter, we used post-classification analysis to evaluate land use land cover changes. The classification accuracy results yielded Overall accuracy of 91.54% and 95.58%, and Kappa Coefficient 0.89 and 0.95 for 2002 and 2014 images respectively. Similar, quantitative analysis of changes expressed in land area coverage reveals a decrease in the thin forest class by 13.7% while the dense forest reduced by 17.3%. Conversely, build-up area appreciated in land coverage by 13.8%.