Urban dispersion-driven land use change detection using GIS-based buffer gradient and Shannon's entropy analysis in...

Rapid and often unregulated urban expansion in ecologically sensitive coastal districts poses challenges to environmental integrity and urban resilience. This study quantifies land use/land cover (LULC) change and urban dispersion in Dakshina Kannada district, India, from 2000 to 2024 using multi-temporal Landsat 7 ETM+ and Landsat 8 OLI imagery, Geographic Information Systems (GIS), and Shannon's entropy. Six LULC categories-agricultural land, forest, built-up land, water bodies, barren land, and sand-were classified using a supervised maximum likelihood approach, achieving overall accuracies > 89% and Kappa coefficients > 0.82 by Error-matrix approach. Results indicate a substantial increase in built-up land (792.9%) over the study period, primarily at the expense of agricultural land, alongside notable forest-agriculture transitions. Shannon's entropy values (0.74-0.77) suggest a moderately dispersed urban form. The 2000-2014 period exhibits higher spatial dispersion and polycentric expansion, while a marginal decline in entropy during 2014-2024 indicates a relative shift toward spatial consolidation. Gradient and buffer-based directional analyses reveal pronounced growth along southeast, south, and northeast corridors aligned with major transport routes, with limited expansion toward the western coastal margin due to ecological and regulatory constraints. Unlike conventional entropy-based studies, this analysis integrates buffer-gradient and directional entropy assessment at the district scale, improving detection of peri-urban and corridor-oriented growth. The findings highlight a transition from dispersed to relatively consolidated urban expansion and underscore the importance of integrated spatial planning to balance development pressures with ecological conservation in rapidly urbanizing coastal regions.