Geospatial assessment of ground deformation in the Greater Accra Region of Ghana using SBAS-InSAR and Sentinel-1 time series

Abstract

The Greater Accra Region of Ghana is experiencing rapid urban expansion, increasing infrastructure loading, and growing hydrological stress, all of which heighten susceptibility to ground deformation. This study integrates multi-temporal Landsat-8 land use/land cover (LULC) mapping for 2017, 2020, and 2024, Sentinel-1 SBAS-InSAR time-series analysis for January 2018 to March 2025, GNSS validation, and an Analytic Hierarchy Process (AHP)-based susceptibility model to characterize deformation patterns and their controlling factors across the region. The LULC results reveal pronounced urban growth, with built-up area increasing from 508.78 km² (15.68%) in 2017 to 850.95 km² (26.22%) in 2024, while bare land declined from 1372.63 km² (42.30%) to 982.54 km² (30.28%) over the same period. SBAS-InSAR results indicate predominantly low-magnitude motion across much of the region, interrupted by six spatially coherent subsidence hotspots covering approximately 573.95 km² (17.69% of the region). Pixel-scale LOS velocities ranged from − 14.23 to + 4.13 mm/yr, while hotspot mean velocities ranged from − 3.76 to − 2.44 mm/yr, with within-zone standard deviations of 1.17–2.80 mm/yr. GNSS-based validation showed close agreement with the SBAS-InSAR time series at the monitored stations, providing independent site-specific support for the satellite-derived deformation estimates. Variance Inflation Factor screening reduced the initial conditioning variables from ten to eight prior to AHP weighting, and the resulting susceptibility map shows that high to very high susceptibility accounts for 1384.37 km² (42.66%) of the region. A quantitative spatial-overlap assessment showed that 66.11% of the valid susceptibility pixels within the six SBAS-InSAR deformation hotspots occurred within the High and Very High susceptibility classes, compared with 42.66% of the mapped region classified as High or Very High susceptibility. This provides quantitative support for the spatial consistency between the observed deformation hotspots and the modelled susceptibility pattern. The principal contribution of the study is an integrated observation–validation–susceptibility framework that links measured SBAS-InSAR deformation, independent GNSS observations, multi-temporal land-surface change and GIS-based susceptibility modelling, providing a transferable basis for deformation screening and risk-informed planning in data-sparse urban environments.