Assessing the impact of urban LULC dynamic on green space in rapidly growing city in eastern India using geospatial techniques

Rapid urbanization and a growing population of over 4.5 million have caused significant changes in land use and land cover (LULC) in Kolkata, leading to the degradation and loss of urban green spaces (UGS), which are important for both the environment and human well-being.This study aims to monitor, analyse, the impact of LULC changes on UGS in Kolkata by integrating geospatial and machine learning (ML) techniques. Multi-temporal Landsat 5 and 8 satellite imagery, enhanced with spectral indices were classified using Support Vector Machine (SVM), Random Forest (RF), and Classification and Regression Trees (CART) within the Google Earth Engine (GEE). Morphological Spatial Pattern Analysis (MSPA) was employed to evaluate the structural transformation in UGS. Additionally, future LULC scenarios for 2031 and 2041 were simulated using Cellular Automata–Artificial Neural Network (CA–ANN) model employed through the MOLUSCE plugin in QGIS. The RF classifier found highest accuracy (98%) with Kappa coefficient of 0.97. From 1991 to 2021, urban impervious surfaces (UIS) increased from 77.17 km² to 123.96 km² (25.10%), largely replacing UGS, which sank from 100.95 km² to 54.12 km² (25.09%). MSPA revealed a noticeable decline in core pattern of UGS from 48.65 km² to 16.19 km², mainly in southern and eastern parts of Kolkata. Further, reduced connectivity in perforation and bridge patterns are observed. Future projections show continuous UIS increase and green space loss, with UIS growing to 128.30 km² and UGS shrinking to 50.64 km² by 2041. The study proposes the implementation of sustainable urban planning policies aimed at preserving and restoring green spaces, promoting urban greening initiatives such as pocket parks, vertical gardens and rooftop greenery, and encouraging public participation to enhance ecological resilience — supporting Sustainable Development Goal (SDG) 11 and SDG 15.