Image Enhancement and Change Detection for Urban Sprawl Analysis of Bauchi Metropolis, Nigeria Using Remote Sensing and GIS Techniques

The objective of this study is to enhance and classify satellite images, and to use them to examine LC/LU changes and urban growth pattern in Bauchi metropolis, Nigeria over a period of three decades (1985-2015). The study utilised multitemporal Landsat satellite images of four different epochs; Thematic Mapper (TM) of 1985 and 1995, Enhanced Thematic Mapper (ETM+) of 2005 and Operational Land Imager (OLI) of 2015. Color composites were made after which geometric and radiometric corrections were performed for the four images using image-to-map and Empirical Line Calibration methods, respectively. Then the images were classified using maximum likelihood algorithm. The classification accuracy was assessed through cross-validation using confusion matrices and field validation through ground truthing with the aid of a hand-held Global Positioning System (GPS). The classified images with their Kappa indexes are; TM of 1985 (0.90%), TM of 1995 (0.82%), ETM+ of 2005 (0.91%), and OLI of 2015 (0.95%). Post-classification comparisons and cross-tabulation analyses were performed and the results showed that changes took place in bare surface (+78.75%), built-up area (+568.77%), farmland (+84.01%), forest (-92.60%), rock outcrop (-56.18%) and water body (+59.78%) between 1985 and 2015 in the metropolis. This shows an unusual change (increase) in water body (lake) over the study period which is largely attributed to sand/rock mining from and around the lake for building and other construction works, in and around the metropolis. It also shows a rapid decrease in forest (vegetation), which may not be unconnected with improper environmental protection strategy in the metropolis. The study demonstrates the relevance of remote sensing and GIS in assessing urban growth as well as revealing the LC and LU situation in the metropolis and the result will serve as an input into a predictive model for urban sprawl analysis.