Coastal spatial planning using object-based image analysis and image classification techniques

Abstract

This study focuses on optimizing coastal spatial planning for Rio de Janeiro, Brazil, addressing critical challenges such as unmanaged urbanization, loss of natural habitats, and land degradation. Utilizing Object-Based Image Analysis (OBIA) and Image Classification techniques within QGIS, the research examines Land Use/Land Cover (LULC) data from 2005, 2015, and 2024. Using the Molusce tool for spatial modeling, land use patterns were predicted for 2035 and 2045. Key spatial parameters analyzed include proximity to rivers, land value, slope level, and population density. Significant trends include a decline in water bodies (3.70% in 2005 to 2.70% in 2045), a surge in built-up areas (29.33%–43.99%), and a reduction in forest cover (38.70%–30.26%). Additionally, barren land is expected to increase from 6.26% to 15.35%, reflecting environmental degradation, while agricultural land shows a steep decline (21.32%–5.59%), driven by urban sprawl and economic shifts. Salt pans, though minimal, are projected to rise from 0.69% to 2.11%. These changes highlight pressing issues like habitat destruction, declining agricultural land, and rising urban pressures in the region. The study underscores the importance of integrated coastal management strategies to balance development and sustainability. Predictive modeling tools in QGIS provide actionable insights for policymakers to anticipate land use changes and implement proactive measures for urban and environmental planning.