<p>This study analyzes shoreline dynamics along the Sfax coastline over a 30-year period using multi-temporal Landsat satellite imagery (TM, ETM⁺, and OLI/TIRS) acquired in the years 1992, 1997, 2002, 2007, 2012, 2017, and 2022. The methodology integrates GIS tools, remote sensing techniques, and the Digital Shoreline Analysis System (DSAS). Following radiometric and geometric corrections, shoreline positions were semi-automatically extracted using the Modified Normalized Difference Water Index (MNDWI). Shoreline change rates were assessed using statistical methods such as End Point Rate (EPR), Net Shoreline Movement (NSM), and Linear Regression Rate (LRR). The coastline was divided into five distinct sectors, Amra-Jbeniana, Sidi Mansour, Casino-Taparura, Chaffar-Mahres, and Skhira analyzed through transects spaced at 50-meter intervals. Results reveal a dominant accretion trend in the Amra-Jbeniana sector, while severe erosion affects Sidi Mansour and Skhira. These patterns reflect increasing vulnerability to both natural factors (sea level rise, wave dynamics, sediment transport) and anthropogenic pressures (e.g., urban development, coastal infrastructure, sand mining). Predictive modeling using the Kalman filter chosen for its ability to assimilate historical trends and update with new data, forecasts continued shoreline retreat in critical sectors, with significant erosion projected by 2032 and worsening by 2042. These findings underscore the urgent need for proactive and sustainable coastal management. Continuous monitoring and targeted interventions are essential to protect the physical integrity and socio-economic value of the Sfax shoreline. This study highlights significant shoreline retreat in the region of sfax and provides valuable insights to enable researchers and decision-makers to better understand regional shoreline dynamics. However, uncertainties related to data availability and seasonal variability should be taken into consideration when interpretating the results.</p> Graphical Abstract <p>The graphical abstract provides a clear overview of the methodology and main findings of this study, which focuses on analyzing the spatio-temporal evolution of the shoreline in the coastal area of the Sfax governorate (Tunisia) over the period 1992–2022. The study area is divided into five sectors. Multi-date Landsat satellite images were selected, preprocessed, and used to compute the Modified Normalized Difference Water Index (MNDWI) to enhance the land-water boundary. An unsupervised k-means classification was then applied to extract the shoreline positions accurately. The extracted shorelines were analyzed using the Digital Shoreline Analysis System (DSAS) to calculate shoreline change rates using the End Point Rate (EPR) and Linear Regression Rate (LRR) methods. This analysis helped identify erosion zones (shown in red) and accretion zones (shown in green), highlighting contrasting coastal dynamics driven by both natural processes and human interventions, particularly the Taparura Project. Finally, a projection of future shoreline positions was conducted for the years 2032 and 2042 using the Kalman filter integrated within DSAS, providing valuable insights for anticipating long-term coastal changes. Overall, the graphical abstract illustrates an integrated remote sensing approach that combines automated extraction, temporal analysis, and predictive modeling in a vulnerable coastal environment.</p>