Delineating Urban Growth Boundary Using Remote sensing, ANN-MLP and CA model: A Case Study of Thiruvananthapuram Urban Agglomeration, India

This research paper investigates urban sprawl in Thiruvananthapuram Urban Agglomeration (UA) and attempts to delineate Urban Growth Boundary (UGB) for promoting urban sustenance. A 112% rise in the spatial expanse of Thiruvananthapuram UA from 256.22 km2 in 2001 to 542.57 km2 in 2011 might induce urban sprawl in the peripheral areas. The Landsat satellite imagery for the years 1987, 1997, 2007, and 2017 were extracted to examine the spatiotemporal urban growth pattern. Shannon’s entropy index was employed to detect urban sprawl in Thiruvananthapuram UA. The UGB delineation process involved future urban growth prediction using the MOLUSCE (Modules for Land Use Change Simulations) plug-in of QGIS software. ANN-MLP (Artificial Neural Network-Multi Layer Perceptron) and CA (Cellular Automata) model was preferred in MOLUSCE to predict future urban growth for the year 2027. Thereafter, hexagons of one square kilometer were used to demarcate the Contiguous built-up Growth Boundary (CGB), and later, sub-administrative units were selected to delineate UGB. The results revealed a rise in the built-up areas from 36.04 km2 in 1987 to 140.69 km2 in 2017. Shannon’s entropy index indicated the prevalence of urban sprawl in Thiruvananthapuram UA. The future growth prediction by 2027 exhibited a further rise in built-up areas to 173.31 km2. The total area within CGB is 213.58 km2, while UGB accounted for 355.59 km2, which included 16 sub-administrative units. This study exhibited a unique methodology to delineate the urban growth boundary, which optimizes the future land requirements in developing nations.