A modelling approach of cellular automata-based artificial neural network for investigating dynamic urban expansion in Kolkata urban agglomeration

Urbanization is a rapidly expanding process in which towns and cities expand vertically and horizontally as a result of increased population density. Thus, one of the most noticeable effects of uncontrolled urban expansion is urban sprawl. As a result, land cover and land use are rapidly changing in cities of developing countries. This article investigates the change in the LU/LC pattern for the period of 1991 and 2021 and simulates the pattern during 2031, 2041, 2051, and 2100 in Kolkata Urban Agglomeration (KUA). It also examines the flow of people inside the Kolkata Urban Agglomeration in order to address the issue of the rapidly increasing rate of urbanization. The mapping of LU/LC patterns from Landsat datasets has been done using the k-means clustering technique. Subsequently, the research used an integrated cellular automata Markov model in combination with TerrSet to predict and model possible land-use/cover scenarios. Using a number of key influencing factors, the study attempts to forecast future urban growth. It has been found in simulated maps that the natural vegetation and agricultural land will continue to decrease between 2021 and 2100.It has also been predicted that the built-up area may be increased from 550.66 km² in 2021 to 682.51 km² in 2100.Relative entropy statistics show that urban sprawl in the KUA has grown faster than expected during the past three decades, which indicates that the log n value was recorded as 0.6021, the highest significant value 0.1579 in 2021, and the lowest value was 0.141 in 1991. Nonetheless, emigration to neighbouring areas has resulted in a decrease in Kolkata's population density. Additional support for the model's applicability as a study field description is provided by the kappa coefficient of 0.8. The study's findings contribute to our understanding of urban sprawl, land-use/land-cover dynamics, and future projections. Furthermore, they offer crucial data for planning and decision-making processes, encouraging sustainable land use management and guiding strategies for prudent urban growth in the study area.