Utilizing Machine Learning and DSAS to Analyze Historical Trends and Forecast Future Shoreline Changes Along the River Niger, Niger Delta

This study investigates shoreline changes along the River Niger in Nigeria over a 70-year period (1974–2044). Using remote sensing data, machine learning, and the DSAS tool, we analyze historical shoreline dynamics from 1974 to 2024 and forecast trends for 2024 to 2044. Data sources include Landsat imagery and SRTM from the United States Geological Survey (USGS) and Sentinel-2 imagery accessed via the Google Earth Engine API. Analysis was conducted using ArcGIS, DSAS versions 5.0 and 6.0, and supplemented with six sediment samples from the riverbank and rainfall data from the Center for Hydrometeorology and Remote Sensing (CHRS). Results reveal significant spatial and temporal variations in shoreline behavior across Bayelsa, Delta, and Anambra states. Approximately 51.47% of transects exhibited erosion, while 48.53% showed accretion, with an average annual shoreline change rate of 1.66 m. Erosion had a greater impact, with a mean rate of − 2.26 m/year, compared to an accretion rate of 3.92 m/year. The total shoreline change envelope (SCE) was 442.86 m, and the net shoreline movement (NSM) was 92.33 m, indicating overall shoreline advancement. Projections for 2024–2044 predict diverse erosion and accretion patterns across different sections, with Section D being the most vulnerable, where 80% of transects are expected to experience erosion at a rate of − 2.96 m/year. Further analysis indicates that the rate and extent of erosion are significantly influenced by the interaction of elevation, slope, and sediment characteristics. Rainfall data analysis shows a strong correlation (R² = 0.7576) between precipitation and shoreline change, underscoring the critical role of climate in coastal dynamics. These findings highlight the urgent need for integrated coastal management strategies that account for rainfall variability and focus on mitigating erosion, particularly in at-risk areas.