MODELLING OF URBAN GROWTH WITH LAND CHANGE MODELER IN OTUKPO METROPOLIS OF BENUE STATE, NIGERIA

The study was aimed at modelling urban growth in Otukpo area of Benue State The study covered a period of 30 years; from 1987 to 2017, and the major transitions to urban were modelled to predict the future scenarios in 2030. Three Landsat satellite images of 1987, 2007 and 2017 were classified using maximum likelihood classifier in Idrisi Selva to detect the land cover changes and a classification accuracy of 84.85%, 85.59% and 86.44% for 1987, 2007 and 2017 maps respectively was achieved. The result of the classification revealed that between 1987 and 2017, urban area gained 12224ha (376.01%) with an annual rate of change of 12.53% while forest lost16493ha (-49.63%) at the rate of -1.65% per year. Physical and proximity factors were identified as major factors driving urban growth in the area. It was found that evidence likelihood of transition, population density the distance from railway and elevation were the most important factors shaping urban growth in the area. Thereafter, a Multilayer Perceptron Markov (MLPMarkov) model was used to model transition potentials of various LULC types to predict future changes. The models had a reliability of 81.7% after validation. The results of the prediction show that Otukpo will experience increase in urban area from 11.59% to 12.6% and forest will decline from 12.54% to 10.98%. It reveals that, Otukpo will grow at the rate of 1.01%. Analysis of the prediction revealed that the rate of urban growth will continue and would certainly threaten other land covers in the area. Keywords— Urban growth Otukpo, Landsat satellite images, maximum likelihood classifier, Idrisi Selva, evidence likelihood of transition, Multilayer Perceptron Markov