Comparison of Spatial Modelling Approaches to Predict Urban Growth of Lucknow City, India

The study presents a comparative analysis of the three urban growth prediction models, i.e., SLEUTH, Markov Chain-Cellular Automata (MC-CA), MC-Multi-layer perceptron (MLP). The capability of these models is assessed for predicting urban growth of Lucknow city, India, for the year 2025. The study resulted in various merits and demerits of the models. The calibration and validation are carried out on remote sensing satellite data of 1995, 2005, 2009 and 2016. The resulted kappa coefficients for CA-MC, MC-MLP and SLEUTH are 0.66, 0.74 and 0.76, respectively. SLEUTH and MC-CA resulted in optimum accuracy, whereas, MC-CA performed most accurately in terms of influential based growth of the city.