Monitoring land use changes and its future prospects using cellular automata simulation and artificial neural network for Ahmedabad city, India

Abstract

The urban sprawl is one of the potential threat for sustainable development. The present study has been carried out on Ahmedabad Municipal Corporation (AMC) area including Ahmedabad city to monitor the urban sprawl from 1976 to 2017. Normalized Difference Vegetation Index, Normalized Difference Built-Up Index, and Built-Up Index are applied to monitor the urban land use change by using multi-temporal and multi-date remote sensing satellite data from Landsat. The future scenario of urban land use prediction of AMC has been modelled by the use of cellular automata (CA) and artificial neural network (ANN) with GIS techniques. The results of land use monitoring indicate that the built-up area has increased by 156.93 km2 whereas, agricultural land, open spaces, urban vegetation, and water bodies have decreased during the last 4-decades from 1976 to 2017. Moreover, the results of CA simulation reveals that the built-up will continue to increase in the future at the cost of other classes. The decadal built-up growth was 2.58 km2 during the earlier period and it was the highest (5.98 km2) during the decade of 2007–2017. The validation of results related to CA–ANN has been measured by the Kappa coefficient in which the obtained values of Kappa local, Kappa histogram and Kappa overall were 0.92, 0.75, and 0.69 respectively and the measured percentage of correctness was 78.63%. The predicted urban land use growth reveals that the built-up would cover the maximum part of the AMC area by 2027 provided that the present land use trend, demographic growth, and commercial development does not show any major change. Therefore, the study pertaining to LULC change and its future prediction will be highly helpful for urban planners and administrators to prepare a sustainable city planning for the Ahmedabad city and its surroundings.