Comparative framework for spatially explicit urban growth modeling for monitoring urban land-use efficiency and sustainable urban development (SDG 11.3.1): a study on Kolkata metropolitan area, India

Abstract

The present study attempts to model the spatiotemporal urban growth of the Kolkata metropolitan area (KMA), India, in a comparative modeling framework using three (remote sensing and geographic information system) RS-GIS integrated techniques, namely stochastic-choice Markov-chain (STCHOICE), cellular automata-Markov (CA-MARKOV), and multi-layer perceptron neural network (MLPNN) coupled with Markov-chain approaches intending to monitor land-use efficiency defined by United Nations for sustainable urban development in KMA. In order to find out the best modeling approach, each of the three techniques is engaged in modeling KMA’s growth, and the model thus obtained is employed to predict future growth. The MLPNN (Kappa = 0.9025) appears to be a considerably better approach as compared to the other two approaches: CA-MARKOV (Kappa = 0.6941) and STCHOICE (Kappa = 0.5392). The MLPNN simulated urban growth for 2036 reveals that the urban built-up cover is expected to be about 55% from that of 31% in 2016 due to the significant conversion of other land covers. The study reveals that urban and peri-urban areas do not have a similar pattern of land consumption and land-use efficiency in KMA. The central KMA reflects better land-use efficiency due to the compact built-up growth compared to the periphery reflecting leapfrog growth as a result of rapid urban sprawling. HIGHLIGHTS Predicted future urban growth in Kolkata Metropolitan Area (KMA) using comparative modeling framework Monitored urban land use efficiency (SDG 11.3.1) for sustainable urban development in KMA The study reveals that urban and peri-urban land consumption and land use efficiency are different Identified most influential driving factors from the selected factors responsible for the spatial pattern of urban expansion in KMA