Proposal and validation of a transferable high-resolution monitoring framework with hotspot analysis for adaptive urban spatial planning in the Global South.

Adaptive spatial planning requires robust spatiotemporal monitoring of urban land use and land cover change, yet no consensus exists on appropriate spatial, temporal, and spectral resolutions for cities in the Global South. This study addresses this gap by developing and validating a high-resolution monitoring framework for urban expansion to support local spatial planning. Using Quito as a case study, impervious surfaces were mapped for 2013, 2017, and 2023, enabling analysis of two five-year intervals. A random forest classifier was implemented using an integrated set of predictors combining vegetation-sensitive spectral indices (NDVI and SAVI), the Perpendicular Impervious Surface Index (PISI), and topographic and textural variables derived from high-resolution multispectral imagery. Classification outputs were integrated with the Getis–Ord G statistic to identify significant spatial clusters of urban expansion. Results show strong classification performance, with consistency values exceeding 91% against independently interpreted reference samples. NDVI and SAVI exhibited the greatest explanatory contribution, while PISI and topographic variables improved discrimination in heterogeneous urban landscapes. The findings support five-year monitoring intervals as an effective balance between change detection and relevance for local spatial planning. Based on these findings, we propose a transferable 5 × 5 × 5 monitoring framework, integrating five-year temporal resolution, multispectral imagery, and 5 m spatial resolution as a baseline for urban expansion monitoring in the Global South, while recognizing that its applicability should be validated across different urban contexts. Hotspot analysis revealed heterogeneous expansion dynamics, providing actionable insights for urban growth management, land-use planning, and sustainable spatial development.