Future land use and land cover simulations with cellular automata-based artificial neural network: A case study over Delhi megacity (India)
Madhavi Jain
DOI: https://doi.org/10.1016/j.heliyon.2024.e34662
IF: 3.776
2024-07-15
Heliyon
Abstract:According to United Nations projections, future global urban growth will mostly occur in Asian megacities. In this study, a Cellular Automata based Artificial Neural Network (CA-ANN) model is used to simulate the future land use and land cover (LULC) over Delhi megacity (India). Delhi, projected to become the world's most populated city by 2030, is an example of a data poor city in Asia, having millions of climate vulnerable people. The CA-ANN model of Modules for Land Change Simulation (MOLUSCE), an open-source plugin, is first tested to simulate the LULC for 2009. Based on good validation results-structural similarity (SSIM; 0.8288), overall accuracy (79.78 %), kappa index of agreement (KIA; 77.25 %), and minimum validation overall error (0.0379), the same model set-up is used to carry out LULC simulation for 2030. This model is found to be simple, efficient, and computationally less expensive tool, and can be used to model future LULCs with a minimal set of inputs, a constraint often found in data poor cities. Results show continued increase in built-up area from 38.3 % (2014) to 53.8 % (2030), at the expense of cultivable areas, forests, and wastelands. The study incorporates past and future LULC change trajectories to highlight the changing LULC dynamics of the megacity from 1977 to 2030. Rate of urban sprawl, calculated using compound annual growth rate (CAGR) is projected to be 2.51 % for 2014-2030, substantially higher than the estimates for 2006-2014 (0.62 %). Further, the past and future urban growth patterns for Delhi are found to mimic other big Asian cities. The database generated from the present study has wide applicability for scientific research community, governmental bodies, profit and non-profit organizations for topics concerning-future urban climate research, climate risk and adaption policy frameworks, climate finance budgeting, future town planning, etc.