Change Detection Analysis of Coastal Mangrove Degradation in Akwa Ibom State Using Multi-Temporal Optical and SAR Data

Mangrove ecosystems in Akwa Ibom State are under sustained degradation pressure from oil pollution, illegal logging, and aquaculture pond conversion, yet spatially accurate, multi-temporal assessments of mangrove change that distinguish degradation type and severity are absent for this state. This study applies multi-temporal optical and SAR data fusion for change detection analysis of mangrove degradation in coastal Akwa Ibom State between 2015 and 2024. Sentinel-2 time-series imagery is processed to derive annual mangrove canopy greenness composites, and Sentinel-1 cross-polarisation SAR data are used to detect structural canopy loss independent of cloud cover. A four-class change detection typology distinguishing persistent mangrove, gradual degradation, acute loss, and partial recovery is developed and applied using a bitemporal image differencing approach with machine learning change classification refinement. Field validation at ninety-two ground reference sites across Eket, Ibeno, and Eastern Obolo LGAs achieves an overall change classification accuracy of 87.1 percent. Results reveal a net mangrove area loss of 6,200 hectares over the study period, with gradual degradation accounting for a larger share of total impact than acute loss events. Oil spill proximity analysis identifies a strong spatial relationship between documented spill events and acute loss polygons. The study contributes spatially explicit degradation data to the Akwa Ibom State Ministry of Environment and the NOSDRA spatial monitoring database. Keywords: mangrove degradation, change detection, SAR, Akwa Ibom, oil pollution.