Identification of urbanization in Ghana based on a discrete approach to analyzing dense Landsat image stacks

Abstract

In this paper a discrete classification approach to land cover and land use changes LCLUC identification based on stable training sites is tested on a nine-date, four year Landsat-7 ETM+ time sequence for a study area in Ghana that is prone to cloud cover. As an indication of urban expansion, change to Built cover was identified for over 70% of testing units when a spatial-temporal majority filter that ignored No Data values from clouds, cloud shadows and sensor effects was applied. Stable LCLU maps were generated and No Data effects should not limit the potential of the approach for longer-term retrospective analyses or monitoring of LCLUC in cloud prone regions.