Spatiotemporal assessment and simulation of land use and land cover dynamics in coastal Tamil Nadu using CA–ANN modelling for sustainable development planning

• Predicted Land Use Land Cover (LULC) changes in the coastal delta region of Tamil Nadu with 88-92% accuracy using the MOLUSCE tool. • By 2053, built-up areas are projected to grow by over 154% and depletion in agricultural and natural lands • Urging species sensitive methods, Normalized Difference Vegetation Index (NDVI) limitations risk misclassifying invasive species and forest recovery • Supporting Sustainable Development Goals (SDG) goals by guiding climate action, sustainable land use and conservation of biodiversity Coastal deltas are vital for food security and biodiversity, but they face unprecedented pressure from rapid urbanization and climate change. However, a critical gap exists in high-resolution, long-term spatial planning for many developing deltas, particularly regarding the misrepresentation of ecological degradation by standard vegetation indices. This study investigates historical Land use and land cover (LULC) patterns and projects future land-use scenarios in the coastal delta taluks of Tamil Nadu, India, using a Cellular Automata–Artificial Neural Network (CA–ANN) model via the MOLUSCE plugin in QGIS. Simulations for the years 2029, 2041, and 2053 were derived from Sentinel-2 satellite data, with map accuracy ranging from 88% to 92%, validated using Kappa statistics (up to 0.972). Findings reveal a significant expansion in built-up areas, a projected increase of over 154% between 2017 and 2053, accompanied by a decline in water bodies, rangelands, and agricultural land. These transformations are primarily driven by elevation, proximity to infrastructure, population growth, and economic development. The study further highlights the limitations of NDVI-based assessments in detecting ecological degradation, particularly due to invasive species being misclassified as forest recovery. By integrating species-sensitive approaches and spatial modelling, this research addresses critical components of several UN Sustainable Development Goals (SDGs) 2, 6, 11, 13 and 15. The results provide a robust framework for policymakers and planners to implement ecologically informed, long-term strategies for landscape resilience, food security, and biodiversity conservation in coastal and climate-sensitive regions.